Benchmarks · /benchmarks/artificial-analysis-output-speed

Output Speed

Name: Output Speed
Creator: Artificial Analysis

Median output generation speed reported by Artificial Analysis.

Source · Artificial Analysis
Version · artificial-analysis snapshot 2026-08-19
Scores · 225

Test details

Visible tradeoffsThis is an efficiency signal, so it belongs beside quality rather than being mistaken for quality.

source

Artificial Analysis

metric

Tokens per second (tokens/s)

judge

Speed / cost

direction

higher better

group id

aa_output_speed_current

domain

Chat / text

What it measures vs what it misses

✓ Measures

Streaming output speed after generation begins.

✗ Misses

Answer quality and time spent reasoning before output starts.

Why this countsIt tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.Same-test ruleThis percentile only compares models inside the exact benchmark/version group shown here. It is not a universal score.What it missesIt does not prove deeper reasoning, tool use, or enterprise workflow reliability.

Leaderboard · this benchmark version

#1 · Celeris-1

AA · Aug 19, 2026

Source label: Celeris-1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 100%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

1,525.9 tokens/s

#2 · mercury-2

AA · Aug 19, 2026

Source label: Mercury 2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 99.6%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

835 tokens/s

#3 · LFM2.5-8B-A1B

AA · Aug 19, 2026

Source label: LFM2.5-8B-A1B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 99.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

338.5 tokens/s

#4 · Gemini 3.5 Flash-Lite

AA · Aug 19, 2026

Source label: Gemini 3.5 Flash-Lite

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 98.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

328.2 tokens/s

#5 · Gemini 3.7 Flash (high)

AA · Aug 19, 2026

Source label: Gemini 3.7 Flash (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 98.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

321.9 tokens/s

#6 · Gemini 3.7 Flash (medium)

AA · Aug 19, 2026

Source label: Gemini 3.7 Flash (medium)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 97.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

320.3 tokens/s

#7 · Nemotron 3 Nano Omni 30B A3B Reasoning

AA · Aug 19, 2026

Source label: Nemotron 3 Nano Omni 30B A3B Reasoning

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 97.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

309.6 tokens/s

#8 · Nemotron 3.5 Lightning

AA · Aug 19, 2026

Source label: Nemotron 3.5 Lightning

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 96.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

303.1 tokens/s

#9 · Gemini 3.1 Flash-Lite Preview

AA · Aug 19, 2026

Source label: Gemini 3.1 Flash-Lite

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 96.4%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

302.7 tokens/s

#10 · Gemini 3.7 Flash (low)

AA · Aug 19, 2026

Source label: Gemini 3.7 Flash (low)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 96%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

282.4 tokens/s

#11 · Nova Micro

AA · Aug 19, 2026

Source label: Nova Micro

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 95.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

282.3 tokens/s

#12 · LFM2.5-VL-1.6B

AA · Aug 19, 2026

Source label: LFM2.5-VL-1.6B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 95.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

277.4 tokens/s

#13 · Qwen3.5 Omni Flash

AA · Aug 19, 2026

Source label: Qwen3.5 Omni Flash

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 94.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

238.5 tokens/s

#14 · Gemini 2.5 Flash-Lite

AA · Aug 19, 2026

Source label: Gemini 2.5 Flash-Lite (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 94.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

231.7 tokens/s

#15 · o3 mini

AA · Aug 19, 2026

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 93.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

221.5 tokens/s

#16 · Step 3.5 Flash

AA · Aug 19, 2026

Source label: Step 3.5 Flash

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 93.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

218.3 tokens/s

#17 · NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)

AA · Aug 19, 2026

Source label: NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 92.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

218 tokens/s

#18 · step-3.5-flash

AA · Aug 19, 2026

Source label: Step 3.5 Flash 2603

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 92.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

213.9 tokens/s

#19 · Ministral 3 3B

AA · Aug 19, 2026

Source label: Ministral 3 3B

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 92%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

203.8 tokens/s

#20 · muse-spark-1.1

AA · Aug 19, 2026

Source label: Muse Spark 1.1 (xhigh)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 91.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

202 tokens/s

#21 · Qwen3.7 Max

AA · Aug 19, 2026

Source label: Qwen3.7 Max

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 91.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

201.4 tokens/s

#22 · Gemini 2.5 Flash

AA · Aug 19, 2026

Source label: Gemini 2.5 Flash (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 90.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

200.3 tokens/s

#23 · Gemini 3.6 Flash (high)

AA · Aug 19, 2026

Source label: Gemini 3.6 Flash (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 90.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

199 tokens/s

#24 · trinity-large-thinking

AA · Aug 19, 2026

Source label: Trinity Large Thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 89.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

190.4 tokens/s

#25 · NVIDIA Nemotron Nano 9B V2 (Reasoning)

AA · Aug 19, 2026

Source label: NVIDIA Nemotron Nano 9B V2 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 89.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

183.4 tokens/s

#26 · qwen3-next-80b-a3b-instruct

AA · Aug 19, 2026

Source label: Qwen3 Next 80B A3B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 88.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

182.6 tokens/s

#27 · Nova 2.0 Lite (medium)

AA · Aug 19, 2026

Source label: Nova 2.0 Lite (medium)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 88.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

169.5 tokens/s

#28 · NVIDIA Nemotron Nano 9B V2 (Non-reasoning)

AA · Aug 19, 2026

Source label: NVIDIA Nemotron Nano 9B V2 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 87.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

168.5 tokens/s

#29 · GPT-OSS 120B

AA · Aug 19, 2026

Source label: gpt-oss-120b (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 87.5%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

163.9 tokens/s

#30 · Nova 2.0 Lite (low)

AA · Aug 19, 2026

Source label: Nova 2.0 Lite (low)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 87.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

163.9 tokens/s

#31 · Gemini 3 Flash

AA · Aug 19, 2026

Source label: Gemini 3 Flash Preview (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 86.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

161 tokens/s

#32 · GPT-5.4 nano

AA · Aug 19, 2026

Source label: GPT-5 nano (medium)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 86.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

158.9 tokens/s

#33 · Mistral Small (Sep '24)

AA · Aug 19, 2026

Source label: Mistral Small (Sep '24)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 85.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

157.6 tokens/s

#34 · Ling 3.0 Tiny

AA · Aug 19, 2026

Source label: Ling 3.0 Tiny

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 85.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

156.3 tokens/s

#35 · Nova Lite

AA · Aug 19, 2026

Source label: Nova Lite

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 84.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

155 tokens/s

#36 · Mistral Small (Feb '24)

AA · Aug 19, 2026

Source label: Mistral Small (Feb '24)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 84.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

154 tokens/s

#37 · Gemini 3.5 Flash

AA · Aug 19, 2026

Source label: Gemini 3.5 Flash (minimal)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 83.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

153.7 tokens/s

#38 · Nova 2.0 Lite (Non-reasoning)

AA · Aug 19, 2026

Source label: Nova 2.0 Lite (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 83.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

152.6 tokens/s

#39 · Nova 2.0 Lite (high)

AA · Aug 19, 2026

Source label: Nova 2.0 Lite (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 83%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

151.9 tokens/s

#40 · o4 mini

AA · Aug 19, 2026

Source label: o4-mini (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 82.6%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

151.9 tokens/s

#41 · Qwen3.5 35B A3B

AA · Aug 19, 2026

Source label: Qwen3.5 35B A3B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 82.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

151.1 tokens/s

#42 · Mistral Small 3.2

AA · Aug 19, 2026

Source label: Mistral Small 3.2

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 81.7%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

150.7 tokens/s

#43 · Solar Pro 3

AA · Aug 19, 2026

Source label: Solar Pro 3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 81.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

149.8 tokens/s

#44 · Mistral Small 3.1

AA · Aug 19, 2026

Source label: Mistral Small 3.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 80.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

149.3 tokens/s

#45 · GPT-5.6 Luna

AA · Aug 19, 2026

Source label: GPT-5.6 Luna (medium)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 80.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

144 tokens/s

#46 · Magistral Small 1.2

AA · Aug 19, 2026

Source label: Magistral Small 1.2

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 79.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

143.3 tokens/s

#47 · Mistral Small 4

AA · Aug 19, 2026

Source label: Mistral Small 4 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 79.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

142.3 tokens/s

#48 · Mistral Small 3

AA · Aug 19, 2026

Source label: Mistral Small 3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 79%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

142.3 tokens/s

#49 · GPT-4.1

AA · Aug 19, 2026

Source label: GPT-4.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 78.6%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

141.8 tokens/s

#50 · Qwen3 30B A3B 2507 Instruct

AA · Aug 19, 2026

Source label: Qwen3 30B A3B 2507 Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 78.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

140.3 tokens/s

#51 · Qwen3.6 35B A3B (Non-reasoning)

AA · Aug 19, 2026

Source label: Qwen3.6 35B A3B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 77.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

139.6 tokens/s

#52 · Nex-N2-Pro

AA · Aug 19, 2026

Source label: Nex-N2-Pro

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 77.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

138 tokens/s

#53 · Qwen3 30B A3B 2507 (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3 30B A3B 2507 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 76.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

137.8 tokens/s

#54 · Llama 3.1 Instruct 8B

AA · Aug 19, 2026

Source label: Llama 3.1 Instruct 8B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 76.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

136.6 tokens/s

#55 · GPT-OSS 20B

AA · Aug 19, 2026

Source label: gpt-oss-20b (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 75.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

135.7 tokens/s

#56 · mistral-medium

AA · Aug 19, 2026

Source label: Mistral Medium

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 75.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

134.8 tokens/s

#57 · Llama 4 Scout

AA · Aug 19, 2026

Source label: Llama 4 Scout

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 75%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

134.2 tokens/s

#58 · Devstral Small 2

AA · Aug 19, 2026

Source label: Devstral Small 2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 74.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

133.2 tokens/s

#59 · Gemini 2.5 Pro

AA · Aug 19, 2026

Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 74.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

133 tokens/s

#60 · Agnes 2.5 Pro Alpha

AA · Aug 19, 2026

Source label: Agnes 2.5 Pro Alpha

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 73.7%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

132 tokens/s

#61 · mistral-medium-3.5

AA · Aug 19, 2026

Source label: Mistral Medium 3.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 73.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

131.9 tokens/s

#62 · Nemotron 3 Ultra 550B A55B (Reasoning)

AA · Aug 19, 2026

Source label: Nemotron 3 Ultra 550B A55B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 72.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

130.9 tokens/s

#63 · Qwen3.5 122B A10B

AA · Aug 19, 2026

Source label: Qwen3.5 122B A10B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 72.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

130.2 tokens/s

#64 · GPT-4.1 nano

AA · Aug 19, 2026

Source label: GPT-4.1 nano

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 71.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

127.6 tokens/s

#65 · nvidia-nemotron-3-super-120b-a12b

AA · Aug 19, 2026

Source label: Nemotron 3 Super 120B A12B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 71.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

127.1 tokens/s

#66 · GPT-5.3 Codex

AA · Aug 19, 2026

Source label: GPT-5.3 Codex (xhigh)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 71%
Last updated: recent
Eligibility: specialized_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

125.1 tokens/s

#67 · Ministral 3 8B

AA · Aug 19, 2026

Source label: Ministral 3 8B

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 70.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

124.8 tokens/s

#68 · Magistral Medium 1.2

AA · Aug 19, 2026

Source label: Magistral Medium 1.2

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 70.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

124.7 tokens/s

#69 · mistral-7b-instruct

AA · Aug 19, 2026

Source label: Mistral 7B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 69.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

123.6 tokens/s

#70 · Qwen3 Coder Next

AA · Aug 19, 2026

Source label: Qwen3 Coder Next

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 69.2%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

122 tokens/s

#71 · Kimi K2 Thinking

AA · Aug 19, 2026

Source label: Kimi K2 Thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 68.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

121.5 tokens/s

#72 · Qwen3.6 35B A3B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3.6 35B A3B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 68.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

118.1 tokens/s

#73 · Step 3.7 Flash

AA · Aug 19, 2026

Source label: Step 3.7 Flash

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 67.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

117.3 tokens/s

#74 · gpt-5.5-instant

AA · Aug 19, 2026

Source label: GPT-5.5 Instant (June 2026)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 67.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

116.7 tokens/s

#75 · granite-4.1-8b

AA · Aug 19, 2026

Source label: Granite 4.1 8B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 67%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

115.3 tokens/s

#76 · Nova 2.0 Pro Preview (medium)

AA · Aug 19, 2026

Source label: Nova 2.0 Pro Preview (medium)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 66.5%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

115.3 tokens/s

#77 · Llama 4 Maverick

AA · Aug 19, 2026

Source label: Llama 4 Maverick

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 66.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

114.6 tokens/s

#78 · Gemma 4 12B (Non-reasoning)

AA · Aug 19, 2026

Source label: Gemma 4 12B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 65.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

114.3 tokens/s

#79 · Qwen3 VL 8B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3 VL 8B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 65.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

113.8 tokens/s

#80 · Muse Glimmer (high)

AA · Aug 19, 2026

Source label: Muse Glimmer (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 64.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

113.1 tokens/s

#81 · o3

AA · Aug 19, 2026

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 64.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

112 tokens/s

#82 · Qwen3 VL 8B Instruct

AA · Aug 19, 2026

Source label: Qwen3 VL 8B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 63.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

109.6 tokens/s

#83 · Gemma 4 12B (Reasoning)

AA · Aug 19, 2026

Source label: Gemma 4 12B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 63.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

109.4 tokens/s

#84 · Llama 3.1 Nemotron Instruct 70B

AA · Aug 19, 2026

Source label: Llama 3.1 Nemotron Instruct 70B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 62.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

108.7 tokens/s

#85 · Gemini 3.1 Pro Preview

AA · Aug 19, 2026

Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 62.5%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

107.9 tokens/s

#86 · Qwen2.5 Turbo

AA · Aug 19, 2026

Source label: Qwen2.5 Turbo

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 62.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

107.7 tokens/s

#87 · Qwen3 VL 30B A3B Instruct

AA · Aug 19, 2026

Source label: Qwen3 VL 30B A3B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 61.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

107.4 tokens/s

#88 · KAT Coder Pro V2

AA · Aug 19, 2026

Source label: KAT Coder Pro V2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 61.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

107.3 tokens/s

#89 · Qwen3 30B A3B (Non-reasoning)

AA · Aug 19, 2026

Source label: Qwen3 30B A3B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 60.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

107.2 tokens/s

#90 · glm-4.7

AA · Aug 19, 2026

Source label: GLM-4.7 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 60.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

106.9 tokens/s

#91 · Mistral Medium 3.1

AA · Aug 19, 2026

Source label: Mistral Medium 3.1

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 59.8%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

106.9 tokens/s

#92 · glm-5.2-max

AA · Aug 19, 2026

Source label: GLM-5.2 (max)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 59.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

105.3 tokens/s

#93 · deepseek-v4-flash

AA · Aug 19, 2026

Source label: DeepSeek V4 Flash 0731 (Reasoning, Max Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 58.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

103.5 tokens/s

#94 · GPT-4.1 mini

AA · Aug 19, 2026

Source label: GPT-4.1 mini

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 58.5%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

103.3 tokens/s

#95 · Qwen3 30B A3B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3 30B A3B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 58%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

102.9 tokens/s

#96 · GPT-4o mini

AA · Aug 19, 2026

Source label: GPT-4o mini

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 57.6%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

101.6 tokens/s

#97 · Qwen3 32B

AA · Aug 19, 2026

Source label: Qwen3 32B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 57.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

101.5 tokens/s

#98 · NVIDIA Nemotron Nano 12B v2 VL (Reasoning)

AA · Aug 19, 2026

Source label: NVIDIA Nemotron Nano 12B v2 VL (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 56.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

100.9 tokens/s

#99 · Nova 2.0 Pro Preview (Non-reasoning)

AA · Aug 19, 2026

Source label: Nova 2.0 Pro Preview (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 56.3%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

99.9 tokens/s

#100 · Qwen3 Omni 30B A3B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3 Omni 30B A3B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 55.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

99.3 tokens/s

#101 · Grok 4.3

AA · Aug 19, 2026

Source label: Grok 4.3 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 55.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

97.9 tokens/s

#102 · GPT-5.1

AA · Aug 19, 2026

Source label: GPT-5.1 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 54.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

95.9 tokens/s

#103 · minimax-m2.5

AA · Aug 19, 2026

Source label: MiniMax-M2.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 54.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

94.5 tokens/s

#104 · GPT-5.6 Terra

AA · Aug 19, 2026

Source label: GPT-5.6 Terra (low)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 54%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

93.7 tokens/s

#105 · Ministral 3 14B

AA · Aug 19, 2026

Source label: Ministral 3 14B

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 53.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

93.6 tokens/s

#106 · Qwen3 Omni 30B A3B Instruct

AA · Aug 19, 2026

Source label: Qwen3 Omni 30B A3B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 53.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

92.9 tokens/s

#107 · glm-4.7-flash

AA · Aug 19, 2026

Source label: GLM-4.7-Flash (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 52.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

92.4 tokens/s

#108 · Ling 2.6 Flash

AA · Aug 19, 2026

Source label: Ling 2.6 Flash

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 52.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

91.9 tokens/s

#109 · Qwen3 Coder 30B A3B Instruct

AA · Aug 19, 2026

Source label: Qwen3 Coder 30B A3B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 51.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

90.5 tokens/s

#110 · GPT-5.4

AA · Aug 19, 2026

Source label: GPT-5.4 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 51.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

90.2 tokens/s

#111 · Claude Haiku 4.5

AA · Aug 19, 2026

Source label: Claude 4.5 Haiku (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 50.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

90.1 tokens/s

#112 · Llama Nemotron Super 49B v1.5 (Non-reasoning)

AA · Aug 19, 2026

Source label: Llama Nemotron Super 49B v1.5 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 50.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

89.8 tokens/s

#113 · Hermes 4 - Llama-3.1 70B (Reasoning)

AA · Aug 19, 2026

Source label: Hermes 4 - Llama-3.1 70B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 50%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

89.5 tokens/s

#114 · minimax-m2

AA · Aug 19, 2026

Source label: MiniMax-M2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 49.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

88.9 tokens/s

#115 · minimax-m3

AA · Aug 19, 2026

Source label: MiniMax-M3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 49.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

88.5 tokens/s

#116 · MiniMax-M2.1

AA · Aug 19, 2026

Source label: MiniMax-M2.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 48.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

88.1 tokens/s

#117 · Claude 4.5 Haiku (Reasoning)

AA · Aug 19, 2026

Source label: Claude 4.5 Haiku (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 48.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

88 tokens/s

#118 · Llama 3.3 Instruct 70B

AA · Aug 19, 2026

Source label: Llama 3.3 Instruct 70B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 47.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

87.6 tokens/s

#119 · Grok 4.20

AA · Aug 19, 2026

Source label: Grok 4.20 0309 v2 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 47.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

87.2 tokens/s

#120 · Qwen3.5 397B A17B

AA · Aug 19, 2026

Source label: Qwen3.5 397B A17B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 46.9%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

84.9 tokens/s

#121 · GLM-5.3 (max)

AA · Aug 19, 2026

Source label: GLM-5.3 (max)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 46.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

84.7 tokens/s

#122 · ling-flash-2.0

AA · Aug 19, 2026

Source label: Ling-flash-2.0

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 46%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

84.5 tokens/s

#123 · glm-4.5v

AA · Aug 19, 2026

Source label: GLM-4.5V (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 45.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

83.4 tokens/s

#124 · GPT-4o

AA · Aug 19, 2026

Source label: GPT-4o (Aug '24)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 45.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

83.4 tokens/s

#125 · glm-5.1

AA · Aug 19, 2026

Source label: GLM-5.1 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 44.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

82.7 tokens/s

#126 · GPT-5.4 mini

AA · Aug 19, 2026

Source label: GPT-5 mini (high)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 44.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

82 tokens/s

#127 · Qwen3.5 9B

AA · Aug 19, 2026

Source label: Qwen3.5 9B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 43.8%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

80.8 tokens/s

#128 · Qwen3.5 27B

AA · Aug 19, 2026

Source label: Qwen3.5 27B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 43.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

77.4 tokens/s

#129 · glm-4.5-air

AA · Aug 19, 2026

Source label: GLM-4.5-Air

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 42.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

76.1 tokens/s

#130 · DeepSeek V4 Pro (Non-reasoning)

AA · Aug 19, 2026

Source label: DeepSeek V4 Pro (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 42.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

76.1 tokens/s

#131 · Claude Sonnet 5 (Adaptive Reasoning, Max Effort)

AA · Aug 19, 2026

Source label: Claude Sonnet 5 (Adaptive Reasoning, Max Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 42%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

76 tokens/s

#132 · Gemma 4 26B A4B

AA · Aug 19, 2026

Source label: Gemma 4 26B A4B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 41.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

74.6 tokens/s

#133 · deepseek-v4-pro

AA · Aug 19, 2026

Source label: DeepSeek V4 Pro (Reasoning, Max Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 41.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

74.1 tokens/s

#134 · glm-4.6v

AA · Aug 19, 2026

Source label: GLM-4.6V (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 40.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

73.2 tokens/s

#135 · Claude Sonnet 5 (Adaptive Reasoning, Xhigh Effort)

AA · Aug 19, 2026

Source label: Claude Sonnet 5 (Adaptive Reasoning, Xhigh Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 40.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

71.4 tokens/s

#136 · Grok 3 mini Reasoning (high)

AA · Aug 19, 2026

Source label: Grok 3 mini Reasoning (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 39.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

70.3 tokens/s

#137 · GPT-5.5

AA · Aug 19, 2026

Source label: GPT-5.5 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 39.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

69.5 tokens/s

#138 · Inkling (xhigh)

AA · Aug 19, 2026

Source label: Inkling (xhigh)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 38.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

68.8 tokens/s

#139 · GPT-5.2

AA · Aug 19, 2026

Source label: GPT-5.2 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 38.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

67.8 tokens/s

#140 · hy3

AA · Aug 19, 2026

Source label: Hy3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 37.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

67.3 tokens/s

#141 · Claude Fable 5

AA · Aug 19, 2026

Source label: Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 37.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

67.1 tokens/s

#142 · Qwen3-Coder 480B A35B

AA · Aug 19, 2026

Source label: Qwen3 Coder 480B A35B Instruct

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 37.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

66.5 tokens/s

#143 · Qwen3 VL 32B Instruct

AA · Aug 19, 2026

Source label: Qwen3 VL 32B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 36.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

66.1 tokens/s

#144 · Inkling Small

AA · Aug 19, 2026

Source label: Inkling Small

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 36.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

65.5 tokens/s

#145 · Qwen3 235B A22B 2507 (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3 235B A22B 2507 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 35.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

64.8 tokens/s

#146 · GPT-5

AA · Aug 19, 2026

Source label: GPT-5 (low)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 35.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

62.9 tokens/s

#147 · claude-sonnet-5-high

AA · Aug 19, 2026

Source label: Claude Sonnet 5 (Adaptive Reasoning, High Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 34.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

62.1 tokens/s

#148 · Qwen3.6 Max Preview

AA · Aug 19, 2026

Source label: Qwen3.6 Max Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 34.4%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

61.8 tokens/s

#149 · GLM-5.1 (Non-reasoning)

AA · Aug 19, 2026

Source label: GLM-5.1 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 33.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

61.4 tokens/s

#150 · minimax-m2.7

AA · Aug 19, 2026

Source label: MiniMax-M2.7

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 33.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

60.6 tokens/s

#151 · Qwen3 235B A22B (Non-reasoning)

AA · Aug 19, 2026

Source label: Qwen3 235B A22B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 33%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

59.5 tokens/s

#152 · Qwen3 14B

AA · Aug 19, 2026

Source label: Qwen3 14B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 32.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

58.9 tokens/s

#153 · Grok Build 0.1 0616

AA · Aug 19, 2026

Source label: Grok Build 0.1 0616

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 32.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

58.9 tokens/s

#154 · Claude Sonnet 5 (Non-reasoning, High Effort)

AA · Aug 19, 2026

Source label: Claude Sonnet 5 (Non-reasoning, High Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 31.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

58.7 tokens/s

#155 · mimo-v2.5

AA · Aug 19, 2026

Source label: MiMo-V2.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 31.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

58.4 tokens/s

#156 · Claude Sonnet 5 (Adaptive Reasoning, Medium Effort)

AA · Aug 19, 2026

Source label: Claude Sonnet 5 (Adaptive Reasoning, Medium Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 30.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

58.1 tokens/s

#157 · Grok 4.6 (high)

AA · Aug 19, 2026

Source label: Grok 4.6 (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 30.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

57.9 tokens/s

#158 · Qwen3 Max

AA · Aug 19, 2026

Source label: Qwen3 Max

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 29.9%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

57.6 tokens/s

#159 · Claude Sonnet 5 (Adaptive Reasoning, Low Effort)

AA · Aug 19, 2026

Source label: Claude Sonnet 5 (Adaptive Reasoning, Low Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 29.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

57.4 tokens/s

#160 · qwen3-235b-a22b-instruct-2507

AA · Aug 19, 2026

Source label: Qwen3 235B A22B 2507 Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 29%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

57.2 tokens/s

#161 · Qwen3 VL 235B A22B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3 VL 235B A22B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 28.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

57.1 tokens/s

#162 · GPT-5.6 Sol

AA · Aug 19, 2026

Source label: GPT-5.6 Sol (low)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 28.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

56.9 tokens/s

#163 · qwen3.7-plus

AA · Aug 19, 2026

Source label: Qwen3.7 Plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 27.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

56.5 tokens/s

#164 · qwen3.6-plus

AA · Aug 19, 2026

Source label: Qwen3.6 Plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 27.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

56.3 tokens/s

#165 · Mistral Medium 3

AA · Aug 19, 2026

Source label: Mistral Medium 3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 26.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

56.3 tokens/s

#166 · Qwen3 Max Thinking

AA · Aug 19, 2026

Source label: Qwen3 Max Thinking (Preview)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 26.3%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

56.2 tokens/s

#167 · Reka Flash (Sep '24)

AA · Aug 19, 2026

Source label: Reka Flash (Sep '24)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 25.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

56 tokens/s

#168 · mimo-v2.5-pro

AA · Aug 19, 2026

Source label: MiMo-V2.5-Pro (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 25.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

55.8 tokens/s

#169 · Gemma 3n E4B Instruct

AA · Aug 19, 2026

Source label: Gemma 3n E4B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 25%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

55.4 tokens/s

#170 · Qwen3.6 27B (Non-reasoning)

AA · Aug 19, 2026

Source label: Qwen3.6 27B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 24.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

55.3 tokens/s

#171 · Command A+

AA · Aug 19, 2026

Source label: Command A

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 24.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

55.2 tokens/s

#172 · Qwen3.6 27B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3.6 27B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 23.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

55.2 tokens/s

#173 · Llama 3.1 Instruct 70B

AA · Aug 19, 2026

Source label: Llama 3.1 Instruct 70B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 23.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

54.8 tokens/s

#174 · Gemma 4 E4B (Non-reasoning)

AA · Aug 19, 2026

Source label: Gemma 4 E4B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 22.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

54.4 tokens/s

#175 · Claude Opus 4.8

AA · Aug 19, 2026

Source label: Claude Opus 4.8 (Adaptive Reasoning, Max Effort)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 22.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

54.4 tokens/s

#176 · Gemma 4 E4B (Reasoning)

AA · Aug 19, 2026

Source label: Gemma 4 E4B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 21.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

54.3 tokens/s

#177 · Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)

AA · Aug 19, 2026

Source label: Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 21.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

53.9 tokens/s

#178 · Claude Opus 5 (Adaptive Reasoning, Max Effort)

AA · Aug 19, 2026

Source label: Claude Opus 5 (Adaptive Reasoning, Max Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 21%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

53 tokens/s

#179 · claude-opus-5-high

AA · Aug 19, 2026

Source label: Claude Opus 5 (Adaptive Reasoning, High Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 20.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

51.5 tokens/s

#180 · Solar Pro 4

AA · Aug 19, 2026

Source label: Solar Pro 4

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 20.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

49.8 tokens/s

#181 · grok-4.5

AA · Aug 19, 2026

Source label: Grok 4.5 (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 19.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

49.8 tokens/s

#182 · glm-4.6

AA · Aug 19, 2026

Source label: GLM-4.6 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 19.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

49.8 tokens/s

#183 · Claude Opus 5 (Adaptive Reasoning, Medium Effort)

AA · Aug 19, 2026

Source label: Claude Opus 5 (Adaptive Reasoning, Medium Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 18.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

49.6 tokens/s

#184 · Qwen3.5 Omni Plus

AA · Aug 19, 2026

Source label: Qwen3.5 Omni Plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 18.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

49.1 tokens/s

#185 · Mistral Large 3

AA · Aug 19, 2026

Source label: Mistral Large 3

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 17.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

48.7 tokens/s

#186 · Claude Opus 5 (Adaptive Reasoning, Low Effort)

AA · Aug 19, 2026

Source label: Claude Opus 5 (Adaptive Reasoning, Low Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 17.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

48.6 tokens/s

#187 · Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)

AA · Aug 19, 2026

Source label: Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 17%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

48 tokens/s

#188 · qwen3-vl-235b-a22b-instruct

AA · Aug 19, 2026

Source label: Qwen3 VL 235B A22B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 16.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

46.9 tokens/s

#189 · Qwen3.8 2.4T A95B

AA · Aug 19, 2026

Source label: Qwen3.8 2.4T A95B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 16.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

46.2 tokens/s

#190 · Llama 3.2 Instruct 11B (Vision)

AA · Aug 19, 2026

Source label: Llama 3.2 Instruct 11B (Vision)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 15.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

46 tokens/s

#191 · glm-5

AA · Aug 19, 2026

Source label: GLM-5 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 15.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

45.6 tokens/s

#192 · Claude Opus 4.5

AA · Aug 19, 2026

Source label: Claude Opus 4.5 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 14.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

44.9 tokens/s

#193 · Qwen3.8 Max

AA · Aug 19, 2026

Source label: Qwen3.8 Max

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 14.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

44.5 tokens/s

#194 · Phi-4 Mini Instruct

AA · Aug 19, 2026

Source label: Phi-4 Mini Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 13.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

43.6 tokens/s

#195 · Kimi K2.5 (Reasoning)

AA · Aug 19, 2026

Source label: Kimi K2.5 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 13.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

43.6 tokens/s

#196 · Claude Opus 4.7

AA · Aug 19, 2026

Source label: Claude Opus 4.7 (Non-reasoning, High Effort)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 12.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

43.6 tokens/s

#197 · Claude Sonnet 4.6

AA · Aug 19, 2026

Source label: Claude Sonnet 4.6 (Non-reasoning, High Effort)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 12.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

43.2 tokens/s

#198 · Hermes 4 - Llama-3.1 70B (Non-reasoning)

AA · Aug 19, 2026

Source label: Hermes 4 - Llama-3.1 70B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 12.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

41.3 tokens/s

#199 · Kimi K2

AA · Aug 19, 2026

Source label: Kimi K2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 11.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

41.2 tokens/s

#200 · Kimi K2 0905

AA · Aug 19, 2026

Source label: Kimi K2 0905

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 11.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

40.9 tokens/s

#201 · kimi-k2.6

AA · Aug 19, 2026

Source label: Kimi K2.6 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 10.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

40 tokens/s

#202 · GLM-5 (Non-reasoning)

AA · Aug 19, 2026

Source label: GLM-5 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 10.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

39.7 tokens/s

#203 · Claude Sonnet 4.5

AA · Aug 19, 2026

Source label: Claude 4.5 Sonnet (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 9.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

38.9 tokens/s

#204 · phi-4

AA · Aug 19, 2026

Source label: Phi-4

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 9.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

38.8 tokens/s

#205 · Qwen3 8B

AA · Aug 19, 2026

Source label: Qwen3 8B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 8.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

38.8 tokens/s

#206 · LongCat 2.0

AA · Aug 19, 2026

Source label: LongCat 2.0

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 8.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

38 tokens/s

#207 · Kimi K3 (max)

AA · Aug 19, 2026

Source label: Kimi K3 (max)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

37.7 tokens/s

#208 · kimi-k2.7-code

AA · Aug 19, 2026

Source label: Kimi K2.7 Code

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 7.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

37.4 tokens/s

#209 · Devstral 2

AA · Aug 19, 2026

Source label: Devstral 2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 7.1%
Last updated: recent
Eligibility: specialized_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

36.8 tokens/s

#210 · Claude Opus 4.6

AA · Aug 19, 2026

Source label: Claude Opus 4.6 (Non-reasoning, High Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 6.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

36.4 tokens/s

#211 · Kimi K3 (low)

AA · Aug 19, 2026

Source label: Kimi K3 (low)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 6.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

36.2 tokens/s

#212 · gemma-4-31b

AA · Aug 19, 2026

Source label: Gemma 4 31B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 5.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

35.9 tokens/s

#213 · Hermes 3 - Llama-3.1 70B

AA · Aug 19, 2026

Source label: Hermes 3 - Llama-3.1 70B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 5.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

33.7 tokens/s

#214 · GPT-4 Turbo

AA · Aug 19, 2026

Source label: GPT-4 Turbo

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 4.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

33.2 tokens/s

#215 · Nova Premier

AA · Aug 19, 2026

Source label: Nova Premier

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 4.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

32.9 tokens/s

#216 · Qwen3.5 4B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3.5 4B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

29.9 tokens/s

#217 · Seed-OSS-36B-Instruct

AA · Aug 19, 2026

Source label: Seed-OSS-36B-Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 3.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

28.5 tokens/s

#218 · Qwen3.5 4B (Non-reasoning)

AA · Aug 19, 2026

Source label: Qwen3.5 4B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 3.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

28.5 tokens/s

#219 · Hermes 4 - Llama-3.1 405B (Reasoning)

AA · Aug 19, 2026

Source label: Hermes 4 - Llama-3.1 405B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 2.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

27.1 tokens/s

#220 · North Mini Code

AA · Aug 19, 2026

Source label: North Mini Code

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 2.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

26.3 tokens/s

#221 · DeepSeek R1 Distill Llama 70B

AA · Aug 19, 2026

Source label: DeepSeek R1 Distill Llama 70B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 1.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

24.8 tokens/s

#222 · Granite 4.0 H Small

AA · Aug 19, 2026

Source label: Granite 4.0 H Small

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 1.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

24.8 tokens/s

#223 · Hermes 4 - Llama-3.1 405B (Non-reasoning)

AA · Aug 19, 2026

Source label: Hermes 4 - Llama-3.1 405B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 0.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

18.7 tokens/s

#224 · Phi-4 Multimodal Instruct

AA · Aug 19, 2026

Source label: Phi-4 Multimodal Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 0.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

17.5 tokens/s

#225 · Granite 3.3 8B (Non-reasoning)

AA · Aug 19, 2026

Source label: Granite 3.3 8B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 0%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

14.4 tokens/s

Benchmarks · /benchmarks/artificial-analysis-output-speed

Output Speed

Median output generation speed reported by Artificial Analysis.

Source · Artificial Analysis
Version · artificial-analysis snapshot 2026-08-19
Scores · 225

Test details

Visible tradeoffsThis is an efficiency signal, so it belongs beside quality rather than being mistaken for quality.

source

Artificial Analysis

metric

Tokens per second (tokens/s)

judge

Speed / cost

direction

higher better

group id

aa_output_speed_current

domain

Chat / text

What it measures vs what it misses

✓ Measures

Streaming output speed after generation begins.

✗ Misses

Answer quality and time spent reasoning before output starts.

Leaderboard · this benchmark version

#1 · Celeris-1

AA · Aug 19, 2026

Source label: Celeris-1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 100%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

1,525.9 tokens/s

#2 · mercury-2

AA · Aug 19, 2026

Source label: Mercury 2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 99.6%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

835 tokens/s

#3 · LFM2.5-8B-A1B

AA · Aug 19, 2026

Source label: LFM2.5-8B-A1B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 99.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

338.5 tokens/s

#4 · Gemini 3.5 Flash-Lite

AA · Aug 19, 2026

Source label: Gemini 3.5 Flash-Lite

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 98.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

328.2 tokens/s

#5 · Gemini 3.7 Flash (high)

AA · Aug 19, 2026

Source label: Gemini 3.7 Flash (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 98.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

321.9 tokens/s

#6 · Gemini 3.7 Flash (medium)

AA · Aug 19, 2026

Source label: Gemini 3.7 Flash (medium)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 97.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

320.3 tokens/s

#7 · Nemotron 3 Nano Omni 30B A3B Reasoning

AA · Aug 19, 2026

Source label: Nemotron 3 Nano Omni 30B A3B Reasoning

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 97.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

309.6 tokens/s

#8 · Nemotron 3.5 Lightning

AA · Aug 19, 2026

Source label: Nemotron 3.5 Lightning

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 96.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

303.1 tokens/s

#9 · Gemini 3.1 Flash-Lite Preview

AA · Aug 19, 2026

Source label: Gemini 3.1 Flash-Lite

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 96.4%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

302.7 tokens/s

#10 · Gemini 3.7 Flash (low)

AA · Aug 19, 2026

Source label: Gemini 3.7 Flash (low)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 96%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

282.4 tokens/s

#11 · Nova Micro

AA · Aug 19, 2026

Source label: Nova Micro

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 95.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

282.3 tokens/s

#12 · LFM2.5-VL-1.6B

AA · Aug 19, 2026

Source label: LFM2.5-VL-1.6B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 95.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

277.4 tokens/s

#13 · Qwen3.5 Omni Flash

AA · Aug 19, 2026

Source label: Qwen3.5 Omni Flash

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 94.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

238.5 tokens/s

#14 · Gemini 2.5 Flash-Lite

AA · Aug 19, 2026

Source label: Gemini 2.5 Flash-Lite (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 94.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

231.7 tokens/s

#15 · o3 mini

AA · Aug 19, 2026

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 93.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

221.5 tokens/s

#16 · Step 3.5 Flash

AA · Aug 19, 2026

Source label: Step 3.5 Flash

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 93.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

218.3 tokens/s

#17 · NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)

AA · Aug 19, 2026

Source label: NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 92.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

218 tokens/s

#18 · step-3.5-flash

AA · Aug 19, 2026

Source label: Step 3.5 Flash 2603

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 92.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

213.9 tokens/s

#19 · Ministral 3 3B

AA · Aug 19, 2026

Source label: Ministral 3 3B

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 92%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

203.8 tokens/s

#20 · muse-spark-1.1

AA · Aug 19, 2026

Source label: Muse Spark 1.1 (xhigh)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 91.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

202 tokens/s

#21 · Qwen3.7 Max

AA · Aug 19, 2026

Source label: Qwen3.7 Max

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 91.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

201.4 tokens/s

#22 · Gemini 2.5 Flash

AA · Aug 19, 2026

Source label: Gemini 2.5 Flash (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 90.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

200.3 tokens/s

#23 · Gemini 3.6 Flash (high)

AA · Aug 19, 2026

Source label: Gemini 3.6 Flash (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 90.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

199 tokens/s

#24 · trinity-large-thinking

AA · Aug 19, 2026

Source label: Trinity Large Thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 89.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

190.4 tokens/s

#25 · NVIDIA Nemotron Nano 9B V2 (Reasoning)

AA · Aug 19, 2026

Source label: NVIDIA Nemotron Nano 9B V2 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 89.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

183.4 tokens/s

#26 · qwen3-next-80b-a3b-instruct

AA · Aug 19, 2026

Source label: Qwen3 Next 80B A3B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 88.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

182.6 tokens/s

#27 · Nova 2.0 Lite (medium)

AA · Aug 19, 2026

Source label: Nova 2.0 Lite (medium)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 88.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

169.5 tokens/s

#28 · NVIDIA Nemotron Nano 9B V2 (Non-reasoning)

AA · Aug 19, 2026

Source label: NVIDIA Nemotron Nano 9B V2 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 87.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

168.5 tokens/s

#29 · GPT-OSS 120B

AA · Aug 19, 2026

Source label: gpt-oss-120b (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 87.5%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

163.9 tokens/s

#30 · Nova 2.0 Lite (low)

AA · Aug 19, 2026

Source label: Nova 2.0 Lite (low)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 87.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

163.9 tokens/s

#31 · Gemini 3 Flash

AA · Aug 19, 2026

Source label: Gemini 3 Flash Preview (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 86.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

161 tokens/s

#32 · GPT-5.4 nano

AA · Aug 19, 2026

Source label: GPT-5 nano (medium)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 86.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

158.9 tokens/s

#33 · Mistral Small (Sep '24)

AA · Aug 19, 2026

Source label: Mistral Small (Sep '24)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 85.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

157.6 tokens/s

#34 · Ling 3.0 Tiny

AA · Aug 19, 2026

Source label: Ling 3.0 Tiny

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 85.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

156.3 tokens/s

#35 · Nova Lite

AA · Aug 19, 2026

Source label: Nova Lite

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 84.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

155 tokens/s

#36 · Mistral Small (Feb '24)

AA · Aug 19, 2026

Source label: Mistral Small (Feb '24)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 84.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

154 tokens/s

#37 · Gemini 3.5 Flash

AA · Aug 19, 2026

Source label: Gemini 3.5 Flash (minimal)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 83.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

153.7 tokens/s

#38 · Nova 2.0 Lite (Non-reasoning)

AA · Aug 19, 2026

Source label: Nova 2.0 Lite (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 83.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

152.6 tokens/s

#39 · Nova 2.0 Lite (high)

AA · Aug 19, 2026

Source label: Nova 2.0 Lite (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 83%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

151.9 tokens/s

#40 · o4 mini

AA · Aug 19, 2026

Source label: o4-mini (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 82.6%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

151.9 tokens/s

#41 · Qwen3.5 35B A3B

AA · Aug 19, 2026

Source label: Qwen3.5 35B A3B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 82.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

151.1 tokens/s

#42 · Mistral Small 3.2

AA · Aug 19, 2026

Source label: Mistral Small 3.2

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 81.7%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

150.7 tokens/s

#43 · Solar Pro 3

AA · Aug 19, 2026

Source label: Solar Pro 3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 81.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

149.8 tokens/s

#44 · Mistral Small 3.1

AA · Aug 19, 2026

Source label: Mistral Small 3.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 80.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

149.3 tokens/s

#45 · GPT-5.6 Luna

AA · Aug 19, 2026

Source label: GPT-5.6 Luna (medium)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 80.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

144 tokens/s

#46 · Magistral Small 1.2

AA · Aug 19, 2026

Source label: Magistral Small 1.2

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 79.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

143.3 tokens/s

#47 · Mistral Small 4

AA · Aug 19, 2026

Source label: Mistral Small 4 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 79.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

142.3 tokens/s

#48 · Mistral Small 3

AA · Aug 19, 2026

Source label: Mistral Small 3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 79%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

142.3 tokens/s

#49 · GPT-4.1

AA · Aug 19, 2026

Source label: GPT-4.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 78.6%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

141.8 tokens/s

#50 · Qwen3 30B A3B 2507 Instruct

AA · Aug 19, 2026

Source label: Qwen3 30B A3B 2507 Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 78.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

140.3 tokens/s

#51 · Qwen3.6 35B A3B (Non-reasoning)

AA · Aug 19, 2026

Source label: Qwen3.6 35B A3B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 77.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

139.6 tokens/s

#52 · Nex-N2-Pro

AA · Aug 19, 2026

Source label: Nex-N2-Pro

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 77.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

138 tokens/s

#53 · Qwen3 30B A3B 2507 (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3 30B A3B 2507 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 76.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

137.8 tokens/s

#54 · Llama 3.1 Instruct 8B

AA · Aug 19, 2026

Source label: Llama 3.1 Instruct 8B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 76.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

136.6 tokens/s

#55 · GPT-OSS 20B

AA · Aug 19, 2026

Source label: gpt-oss-20b (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 75.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

135.7 tokens/s

#56 · mistral-medium

AA · Aug 19, 2026

Source label: Mistral Medium

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 75.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

134.8 tokens/s

#57 · Llama 4 Scout

AA · Aug 19, 2026

Source label: Llama 4 Scout

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 75%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

134.2 tokens/s

#58 · Devstral Small 2

AA · Aug 19, 2026

Source label: Devstral Small 2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 74.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

133.2 tokens/s

#59 · Gemini 2.5 Pro

AA · Aug 19, 2026

Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 74.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

133 tokens/s

#60 · Agnes 2.5 Pro Alpha

AA · Aug 19, 2026

Source label: Agnes 2.5 Pro Alpha

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 73.7%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

132 tokens/s

#61 · mistral-medium-3.5

AA · Aug 19, 2026

Source label: Mistral Medium 3.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 73.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

131.9 tokens/s

#62 · Nemotron 3 Ultra 550B A55B (Reasoning)

AA · Aug 19, 2026

Source label: Nemotron 3 Ultra 550B A55B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 72.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

130.9 tokens/s

#63 · Qwen3.5 122B A10B

AA · Aug 19, 2026

Source label: Qwen3.5 122B A10B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 72.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

130.2 tokens/s

#64 · GPT-4.1 nano

AA · Aug 19, 2026

Source label: GPT-4.1 nano

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 71.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

127.6 tokens/s

#65 · nvidia-nemotron-3-super-120b-a12b

AA · Aug 19, 2026

Source label: Nemotron 3 Super 120B A12B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 71.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

127.1 tokens/s

#66 · GPT-5.3 Codex

AA · Aug 19, 2026

Source label: GPT-5.3 Codex (xhigh)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 71%
Last updated: recent
Eligibility: specialized_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

125.1 tokens/s

#67 · Ministral 3 8B

AA · Aug 19, 2026

Source label: Ministral 3 8B

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 70.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

124.8 tokens/s

#68 · Magistral Medium 1.2

AA · Aug 19, 2026

Source label: Magistral Medium 1.2

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 70.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

124.7 tokens/s

#69 · mistral-7b-instruct

AA · Aug 19, 2026

Source label: Mistral 7B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 69.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

123.6 tokens/s

#70 · Qwen3 Coder Next

AA · Aug 19, 2026

Source label: Qwen3 Coder Next

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 69.2%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

122 tokens/s

#71 · Kimi K2 Thinking

AA · Aug 19, 2026

Source label: Kimi K2 Thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 68.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

121.5 tokens/s

#72 · Qwen3.6 35B A3B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3.6 35B A3B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 68.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

118.1 tokens/s

#73 · Step 3.7 Flash

AA · Aug 19, 2026

Source label: Step 3.7 Flash

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 67.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

117.3 tokens/s

#74 · gpt-5.5-instant

AA · Aug 19, 2026

Source label: GPT-5.5 Instant (June 2026)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 67.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

116.7 tokens/s

#75 · granite-4.1-8b

AA · Aug 19, 2026

Source label: Granite 4.1 8B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 67%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

115.3 tokens/s

#76 · Nova 2.0 Pro Preview (medium)

AA · Aug 19, 2026

Source label: Nova 2.0 Pro Preview (medium)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 66.5%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

115.3 tokens/s

#77 · Llama 4 Maverick

AA · Aug 19, 2026

Source label: Llama 4 Maverick

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 66.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

114.6 tokens/s

#78 · Gemma 4 12B (Non-reasoning)

AA · Aug 19, 2026

Source label: Gemma 4 12B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 65.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

114.3 tokens/s

#79 · Qwen3 VL 8B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3 VL 8B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 65.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

113.8 tokens/s

#80 · Muse Glimmer (high)

AA · Aug 19, 2026

Source label: Muse Glimmer (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 64.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

113.1 tokens/s

#81 · o3

AA · Aug 19, 2026

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 64.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

112 tokens/s

#82 · Qwen3 VL 8B Instruct

AA · Aug 19, 2026

Source label: Qwen3 VL 8B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 63.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

109.6 tokens/s

#83 · Gemma 4 12B (Reasoning)

AA · Aug 19, 2026

Source label: Gemma 4 12B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 63.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

109.4 tokens/s

#84 · Llama 3.1 Nemotron Instruct 70B

AA · Aug 19, 2026

Source label: Llama 3.1 Nemotron Instruct 70B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 62.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

108.7 tokens/s

#85 · Gemini 3.1 Pro Preview

AA · Aug 19, 2026

Source label: Gemini 3.1 Pro Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 62.5%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

107.9 tokens/s

#86 · Qwen2.5 Turbo

AA · Aug 19, 2026

Source label: Qwen2.5 Turbo

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 62.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

107.7 tokens/s

#87 · Qwen3 VL 30B A3B Instruct

AA · Aug 19, 2026

Source label: Qwen3 VL 30B A3B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 61.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

107.4 tokens/s

#88 · KAT Coder Pro V2

AA · Aug 19, 2026

Source label: KAT Coder Pro V2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 61.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

107.3 tokens/s

#89 · Qwen3 30B A3B (Non-reasoning)

AA · Aug 19, 2026

Source label: Qwen3 30B A3B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 60.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

107.2 tokens/s

#90 · glm-4.7

AA · Aug 19, 2026

Source label: GLM-4.7 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 60.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

106.9 tokens/s

#91 · Mistral Medium 3.1

AA · Aug 19, 2026

Source label: Mistral Medium 3.1

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 59.8%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

106.9 tokens/s

#92 · glm-5.2-max

AA · Aug 19, 2026

Source label: GLM-5.2 (max)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 59.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

105.3 tokens/s

#93 · deepseek-v4-flash

AA · Aug 19, 2026

Source label: DeepSeek V4 Flash 0731 (Reasoning, Max Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 58.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

103.5 tokens/s

#94 · GPT-4.1 mini

AA · Aug 19, 2026

Source label: GPT-4.1 mini

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 58.5%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

103.3 tokens/s

#95 · Qwen3 30B A3B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3 30B A3B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 58%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

102.9 tokens/s

#96 · GPT-4o mini

AA · Aug 19, 2026

Source label: GPT-4o mini

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 57.6%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

101.6 tokens/s

#97 · Qwen3 32B

AA · Aug 19, 2026

Source label: Qwen3 32B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 57.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

101.5 tokens/s

#98 · NVIDIA Nemotron Nano 12B v2 VL (Reasoning)

AA · Aug 19, 2026

Source label: NVIDIA Nemotron Nano 12B v2 VL (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 56.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

100.9 tokens/s

#99 · Nova 2.0 Pro Preview (Non-reasoning)

AA · Aug 19, 2026

Source label: Nova 2.0 Pro Preview (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 56.3%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

99.9 tokens/s

#100 · Qwen3 Omni 30B A3B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3 Omni 30B A3B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 55.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

99.3 tokens/s

#101 · Grok 4.3

AA · Aug 19, 2026

Source label: Grok 4.3 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 55.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

97.9 tokens/s

#102 · GPT-5.1

AA · Aug 19, 2026

Source label: GPT-5.1 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 54.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

95.9 tokens/s

#103 · minimax-m2.5

AA · Aug 19, 2026

Source label: MiniMax-M2.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 54.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

94.5 tokens/s

#104 · GPT-5.6 Terra

AA · Aug 19, 2026

Source label: GPT-5.6 Terra (low)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 54%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

93.7 tokens/s

#105 · Ministral 3 14B

AA · Aug 19, 2026

Source label: Ministral 3 14B

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 53.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

93.6 tokens/s

#106 · Qwen3 Omni 30B A3B Instruct

AA · Aug 19, 2026

Source label: Qwen3 Omni 30B A3B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 53.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

92.9 tokens/s

#107 · glm-4.7-flash

AA · Aug 19, 2026

Source label: GLM-4.7-Flash (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 52.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

92.4 tokens/s

#108 · Ling 2.6 Flash

AA · Aug 19, 2026

Source label: Ling 2.6 Flash

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 52.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

91.9 tokens/s

#109 · Qwen3 Coder 30B A3B Instruct

AA · Aug 19, 2026

Source label: Qwen3 Coder 30B A3B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 51.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

90.5 tokens/s

#110 · GPT-5.4

AA · Aug 19, 2026

Source label: GPT-5.4 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 51.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

90.2 tokens/s

#111 · Claude Haiku 4.5

AA · Aug 19, 2026

Source label: Claude 4.5 Haiku (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 50.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

90.1 tokens/s

#112 · Llama Nemotron Super 49B v1.5 (Non-reasoning)

AA · Aug 19, 2026

Source label: Llama Nemotron Super 49B v1.5 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 50.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

89.8 tokens/s

#113 · Hermes 4 - Llama-3.1 70B (Reasoning)

AA · Aug 19, 2026

Source label: Hermes 4 - Llama-3.1 70B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 50%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

89.5 tokens/s

#114 · minimax-m2

AA · Aug 19, 2026

Source label: MiniMax-M2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 49.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

88.9 tokens/s

#115 · minimax-m3

AA · Aug 19, 2026

Source label: MiniMax-M3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 49.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

88.5 tokens/s

#116 · MiniMax-M2.1

AA · Aug 19, 2026

Source label: MiniMax-M2.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 48.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

88.1 tokens/s

#117 · Claude 4.5 Haiku (Reasoning)

AA · Aug 19, 2026

Source label: Claude 4.5 Haiku (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 48.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

88 tokens/s

#118 · Llama 3.3 Instruct 70B

AA · Aug 19, 2026

Source label: Llama 3.3 Instruct 70B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 47.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

87.6 tokens/s

#119 · Grok 4.20

AA · Aug 19, 2026

Source label: Grok 4.20 0309 v2 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 47.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

87.2 tokens/s

#120 · Qwen3.5 397B A17B

AA · Aug 19, 2026

Source label: Qwen3.5 397B A17B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 46.9%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

84.9 tokens/s

#121 · GLM-5.3 (max)

AA · Aug 19, 2026

Source label: GLM-5.3 (max)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 46.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

84.7 tokens/s

#122 · ling-flash-2.0

AA · Aug 19, 2026

Source label: Ling-flash-2.0

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 46%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

84.5 tokens/s

#123 · glm-4.5v

AA · Aug 19, 2026

Source label: GLM-4.5V (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 45.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

83.4 tokens/s

#124 · GPT-4o

AA · Aug 19, 2026

Source label: GPT-4o (Aug '24)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 45.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

83.4 tokens/s

#125 · glm-5.1

AA · Aug 19, 2026

Source label: GLM-5.1 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 44.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

82.7 tokens/s

#126 · GPT-5.4 mini

AA · Aug 19, 2026

Source label: GPT-5 mini (high)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 44.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

82 tokens/s

#127 · Qwen3.5 9B

AA · Aug 19, 2026

Source label: Qwen3.5 9B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 43.8%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

80.8 tokens/s

#128 · Qwen3.5 27B

AA · Aug 19, 2026

Source label: Qwen3.5 27B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 43.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

77.4 tokens/s

#129 · glm-4.5-air

AA · Aug 19, 2026

Source label: GLM-4.5-Air

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 42.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

76.1 tokens/s

#130 · DeepSeek V4 Pro (Non-reasoning)

AA · Aug 19, 2026

Source label: DeepSeek V4 Pro (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 42.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

76.1 tokens/s

#131 · Claude Sonnet 5 (Adaptive Reasoning, Max Effort)

AA · Aug 19, 2026

Source label: Claude Sonnet 5 (Adaptive Reasoning, Max Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 42%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

76 tokens/s

#132 · Gemma 4 26B A4B

AA · Aug 19, 2026

Source label: Gemma 4 26B A4B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 41.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

74.6 tokens/s

#133 · deepseek-v4-pro

AA · Aug 19, 2026

Source label: DeepSeek V4 Pro (Reasoning, Max Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 41.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

74.1 tokens/s

#134 · glm-4.6v

AA · Aug 19, 2026

Source label: GLM-4.6V (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 40.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

73.2 tokens/s

#135 · Claude Sonnet 5 (Adaptive Reasoning, Xhigh Effort)

AA · Aug 19, 2026

Source label: Claude Sonnet 5 (Adaptive Reasoning, Xhigh Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 40.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

71.4 tokens/s

#136 · Grok 3 mini Reasoning (high)

AA · Aug 19, 2026

Source label: Grok 3 mini Reasoning (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 39.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

70.3 tokens/s

#137 · GPT-5.5

AA · Aug 19, 2026

Source label: GPT-5.5 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 39.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

69.5 tokens/s

#138 · Inkling (xhigh)

AA · Aug 19, 2026

Source label: Inkling (xhigh)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 38.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

68.8 tokens/s

#139 · GPT-5.2

AA · Aug 19, 2026

Source label: GPT-5.2 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 38.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

67.8 tokens/s

#140 · hy3

AA · Aug 19, 2026

Source label: Hy3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 37.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

67.3 tokens/s

#141 · Claude Fable 5

AA · Aug 19, 2026

Source label: Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 37.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

67.1 tokens/s

#142 · Qwen3-Coder 480B A35B

AA · Aug 19, 2026

Source label: Qwen3 Coder 480B A35B Instruct

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 37.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

66.5 tokens/s

#143 · Qwen3 VL 32B Instruct

AA · Aug 19, 2026

Source label: Qwen3 VL 32B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 36.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

66.1 tokens/s

#144 · Inkling Small

AA · Aug 19, 2026

Source label: Inkling Small

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 36.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

65.5 tokens/s

#145 · Qwen3 235B A22B 2507 (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3 235B A22B 2507 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 35.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

64.8 tokens/s

#146 · GPT-5

AA · Aug 19, 2026

Source label: GPT-5 (low)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 35.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

62.9 tokens/s

#147 · claude-sonnet-5-high

AA · Aug 19, 2026

Source label: Claude Sonnet 5 (Adaptive Reasoning, High Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 34.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

62.1 tokens/s

#148 · Qwen3.6 Max Preview

AA · Aug 19, 2026

Source label: Qwen3.6 Max Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 34.4%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

61.8 tokens/s

#149 · GLM-5.1 (Non-reasoning)

AA · Aug 19, 2026

Source label: GLM-5.1 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 33.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

61.4 tokens/s

#150 · minimax-m2.7

AA · Aug 19, 2026

Source label: MiniMax-M2.7

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 33.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

60.6 tokens/s

#151 · Qwen3 235B A22B (Non-reasoning)

AA · Aug 19, 2026

Source label: Qwen3 235B A22B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 33%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

59.5 tokens/s

#152 · Qwen3 14B

AA · Aug 19, 2026

Source label: Qwen3 14B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 32.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

58.9 tokens/s

#153 · Grok Build 0.1 0616

AA · Aug 19, 2026

Source label: Grok Build 0.1 0616

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 32.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

58.9 tokens/s

#154 · Claude Sonnet 5 (Non-reasoning, High Effort)

AA · Aug 19, 2026

Source label: Claude Sonnet 5 (Non-reasoning, High Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 31.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

58.7 tokens/s

#155 · mimo-v2.5

AA · Aug 19, 2026

Source label: MiMo-V2.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 31.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

58.4 tokens/s

#156 · Claude Sonnet 5 (Adaptive Reasoning, Medium Effort)

AA · Aug 19, 2026

Source label: Claude Sonnet 5 (Adaptive Reasoning, Medium Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 30.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

58.1 tokens/s

#157 · Grok 4.6 (high)

AA · Aug 19, 2026

Source label: Grok 4.6 (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 30.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

57.9 tokens/s

#158 · Qwen3 Max

AA · Aug 19, 2026

Source label: Qwen3 Max

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 29.9%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

57.6 tokens/s

#159 · Claude Sonnet 5 (Adaptive Reasoning, Low Effort)

AA · Aug 19, 2026

Source label: Claude Sonnet 5 (Adaptive Reasoning, Low Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 29.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

57.4 tokens/s

#160 · qwen3-235b-a22b-instruct-2507

AA · Aug 19, 2026

Source label: Qwen3 235B A22B 2507 Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 29%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

57.2 tokens/s

#161 · Qwen3 VL 235B A22B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3 VL 235B A22B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 28.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

57.1 tokens/s

#162 · GPT-5.6 Sol

AA · Aug 19, 2026

Source label: GPT-5.6 Sol (low)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 28.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

56.9 tokens/s

#163 · qwen3.7-plus

AA · Aug 19, 2026

Source label: Qwen3.7 Plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 27.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

56.5 tokens/s

#164 · qwen3.6-plus

AA · Aug 19, 2026

Source label: Qwen3.6 Plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 27.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

56.3 tokens/s

#165 · Mistral Medium 3

AA · Aug 19, 2026

Source label: Mistral Medium 3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 26.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

56.3 tokens/s

#166 · Qwen3 Max Thinking

AA · Aug 19, 2026

Source label: Qwen3 Max Thinking (Preview)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 26.3%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

56.2 tokens/s

#167 · Reka Flash (Sep '24)

AA · Aug 19, 2026

Source label: Reka Flash (Sep '24)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 25.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

56 tokens/s

#168 · mimo-v2.5-pro

AA · Aug 19, 2026

Source label: MiMo-V2.5-Pro (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 25.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

55.8 tokens/s

#169 · Gemma 3n E4B Instruct

AA · Aug 19, 2026

Source label: Gemma 3n E4B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 25%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

55.4 tokens/s

#170 · Qwen3.6 27B (Non-reasoning)

AA · Aug 19, 2026

Source label: Qwen3.6 27B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 24.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

55.3 tokens/s

#171 · Command A+

AA · Aug 19, 2026

Source label: Command A

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 24.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

55.2 tokens/s

#172 · Qwen3.6 27B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3.6 27B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 23.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

55.2 tokens/s

#173 · Llama 3.1 Instruct 70B

AA · Aug 19, 2026

Source label: Llama 3.1 Instruct 70B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 23.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

54.8 tokens/s

#174 · Gemma 4 E4B (Non-reasoning)

AA · Aug 19, 2026

Source label: Gemma 4 E4B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 22.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

54.4 tokens/s

#175 · Claude Opus 4.8

AA · Aug 19, 2026

Source label: Claude Opus 4.8 (Adaptive Reasoning, Max Effort)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 22.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

54.4 tokens/s

#176 · Gemma 4 E4B (Reasoning)

AA · Aug 19, 2026

Source label: Gemma 4 E4B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 21.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

54.3 tokens/s

#177 · Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)

AA · Aug 19, 2026

Source label: Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 21.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

53.9 tokens/s

#178 · Claude Opus 5 (Adaptive Reasoning, Max Effort)

AA · Aug 19, 2026

Source label: Claude Opus 5 (Adaptive Reasoning, Max Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 21%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

53 tokens/s

#179 · claude-opus-5-high

AA · Aug 19, 2026

Source label: Claude Opus 5 (Adaptive Reasoning, High Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 20.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

51.5 tokens/s

#180 · Solar Pro 4

AA · Aug 19, 2026

Source label: Solar Pro 4

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 20.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

49.8 tokens/s

#181 · grok-4.5

AA · Aug 19, 2026

Source label: Grok 4.5 (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 19.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

49.8 tokens/s

#182 · glm-4.6

AA · Aug 19, 2026

Source label: GLM-4.6 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 19.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

49.8 tokens/s

#183 · Claude Opus 5 (Adaptive Reasoning, Medium Effort)

AA · Aug 19, 2026

Source label: Claude Opus 5 (Adaptive Reasoning, Medium Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 18.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

49.6 tokens/s

#184 · Qwen3.5 Omni Plus

AA · Aug 19, 2026

Source label: Qwen3.5 Omni Plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 18.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

49.1 tokens/s

#185 · Mistral Large 3

AA · Aug 19, 2026

Source label: Mistral Large 3

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 17.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

48.7 tokens/s

#186 · Claude Opus 5 (Adaptive Reasoning, Low Effort)

AA · Aug 19, 2026

Source label: Claude Opus 5 (Adaptive Reasoning, Low Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 17.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

48.6 tokens/s

#187 · Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)

AA · Aug 19, 2026

Source label: Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 17%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

48 tokens/s

#188 · qwen3-vl-235b-a22b-instruct

AA · Aug 19, 2026

Source label: Qwen3 VL 235B A22B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 16.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

46.9 tokens/s

#189 · Qwen3.8 2.4T A95B

AA · Aug 19, 2026

Source label: Qwen3.8 2.4T A95B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 16.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

46.2 tokens/s

#190 · Llama 3.2 Instruct 11B (Vision)

AA · Aug 19, 2026

Source label: Llama 3.2 Instruct 11B (Vision)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 15.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

46 tokens/s

#191 · glm-5

AA · Aug 19, 2026

Source label: GLM-5 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 15.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

45.6 tokens/s

#192 · Claude Opus 4.5

AA · Aug 19, 2026

Source label: Claude Opus 4.5 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 14.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

44.9 tokens/s

#193 · Qwen3.8 Max

AA · Aug 19, 2026

Source label: Qwen3.8 Max

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 14.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

44.5 tokens/s

#194 · Phi-4 Mini Instruct

AA · Aug 19, 2026

Source label: Phi-4 Mini Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 13.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

43.6 tokens/s

#195 · Kimi K2.5 (Reasoning)

AA · Aug 19, 2026

Source label: Kimi K2.5 (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 13.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

43.6 tokens/s

#196 · Claude Opus 4.7

AA · Aug 19, 2026

Source label: Claude Opus 4.7 (Non-reasoning, High Effort)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 12.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

43.6 tokens/s

#197 · Claude Sonnet 4.6

AA · Aug 19, 2026

Source label: Claude Sonnet 4.6 (Non-reasoning, High Effort)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 12.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

43.2 tokens/s

#198 · Hermes 4 - Llama-3.1 70B (Non-reasoning)

AA · Aug 19, 2026

Source label: Hermes 4 - Llama-3.1 70B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 12.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

41.3 tokens/s

#199 · Kimi K2

AA · Aug 19, 2026

Source label: Kimi K2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 11.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

41.2 tokens/s

#200 · Kimi K2 0905

AA · Aug 19, 2026

Source label: Kimi K2 0905

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 11.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

40.9 tokens/s

#201 · kimi-k2.6

AA · Aug 19, 2026

Source label: Kimi K2.6 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 10.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

40 tokens/s

#202 · GLM-5 (Non-reasoning)

AA · Aug 19, 2026

Source label: GLM-5 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 10.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

39.7 tokens/s

#203 · Claude Sonnet 4.5

AA · Aug 19, 2026

Source label: Claude 4.5 Sonnet (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 9.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

38.9 tokens/s

#204 · phi-4

AA · Aug 19, 2026

Source label: Phi-4

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 9.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

38.8 tokens/s

#205 · Qwen3 8B

AA · Aug 19, 2026

Source label: Qwen3 8B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 8.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

38.8 tokens/s

#206 · LongCat 2.0

AA · Aug 19, 2026

Source label: LongCat 2.0

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 8.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

38 tokens/s

#207 · Kimi K3 (max)

AA · Aug 19, 2026

Source label: Kimi K3 (max)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

37.7 tokens/s

#208 · kimi-k2.7-code

AA · Aug 19, 2026

Source label: Kimi K2.7 Code

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 7.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

37.4 tokens/s

#209 · Devstral 2

AA · Aug 19, 2026

Source label: Devstral 2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 7.1%
Last updated: recent
Eligibility: specialized_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

36.8 tokens/s

#210 · Claude Opus 4.6

AA · Aug 19, 2026

Source label: Claude Opus 4.6 (Non-reasoning, High Effort)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 6.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

36.4 tokens/s

#211 · Kimi K3 (low)

AA · Aug 19, 2026

Source label: Kimi K3 (low)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 6.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

36.2 tokens/s

#212 · gemma-4-31b

AA · Aug 19, 2026

Source label: Gemma 4 31B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 5.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

35.9 tokens/s

#213 · Hermes 3 - Llama-3.1 70B

AA · Aug 19, 2026

Source label: Hermes 3 - Llama-3.1 70B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 5.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

33.7 tokens/s

#214 · GPT-4 Turbo

AA · Aug 19, 2026

Source label: GPT-4 Turbo

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 4.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

33.2 tokens/s

#215 · Nova Premier

AA · Aug 19, 2026

Source label: Nova Premier

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 4.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

32.9 tokens/s

#216 · Qwen3.5 4B (Reasoning)

AA · Aug 19, 2026

Source label: Qwen3.5 4B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

29.9 tokens/s

#217 · Seed-OSS-36B-Instruct

AA · Aug 19, 2026

Source label: Seed-OSS-36B-Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 3.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

28.5 tokens/s

#218 · Qwen3.5 4B (Non-reasoning)

AA · Aug 19, 2026

Source label: Qwen3.5 4B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 3.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

28.5 tokens/s

#219 · Hermes 4 - Llama-3.1 405B (Reasoning)

AA · Aug 19, 2026

Source label: Hermes 4 - Llama-3.1 405B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 2.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

27.1 tokens/s

#220 · North Mini Code

AA · Aug 19, 2026

Source label: North Mini Code

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 2.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

26.3 tokens/s

#221 · DeepSeek R1 Distill Llama 70B

AA · Aug 19, 2026

Source label: DeepSeek R1 Distill Llama 70B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 1.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

24.8 tokens/s

#222 · Granite 4.0 H Small

AA · Aug 19, 2026

Source label: Granite 4.0 H Small

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 1.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

24.8 tokens/s

#223 · Hermes 4 - Llama-3.1 405B (Non-reasoning)

AA · Aug 19, 2026

Source label: Hermes 4 - Llama-3.1 405B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 0.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

18.7 tokens/s

#224 · Phi-4 Multimodal Instruct

AA · Aug 19, 2026

Source label: Phi-4 Multimodal Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 0.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

17.5 tokens/s

#225 · Granite 3.3 8B (Non-reasoning)

AA · Aug 19, 2026

Source label: Granite 3.3 8B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 0%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

14.4 tokens/s

Output Speed

Test details

What it measures vs what it misses

✓ Measures

✗ Misses

Leaderboard · this benchmark version

Loading benchmark evidence.

Output Speed

Test details

What it measures vs what it misses

✓ Measures

✗ Misses

Leaderboard · this benchmark version