Benchmarks · /benchmarks/artificial-analysis-mmmu-pro

MMMU-Pro

Name: MMMU-Pro
Creator: Artificial Analysis

MMMU-Pro result as reported through Artificial Analysis.

Source · Artificial Analysis
Version · artificial-analysis snapshot 2026-06-24
Scores · 136

Test details

Visible tradeoffsThis is an objective signal, so it is mainly about measurable task performance rather than public taste.

source

Artificial Analysis

metric

Score (%)

judge

Objective

direction

higher better

group id

aa_mmmu_pro_current

domain

Vision understanding

What it measures vs what it misses

✓ Measures

Multimodal reasoning over images and prompts.

✗ Misses

Adjacent capabilities, subjective preference, latency, and cost.

Why this countsIt is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.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 tell you whether the model can generate or edit images well.

Leaderboard · this benchmark version

#1 · Gemini 3.1 Pro Preview

AA · Jun 24, 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: 100%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

82.4%

#2 · muse-spark

AA · Jun 24, 2026

Source label: Muse Spark

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 99.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

80.5%

#3 · Qwen3.7 Plus

AA · Jun 24, 2026

Source label: Qwen3.7 Plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 98.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

80.5%

#4 · Gemini 3 Pro Preview

AA · Jun 24, 2026

Source label: Gemini 3 Pro Preview (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

80.2%

#5 · Gemini 3.5 Flash

AA · Jun 24, 2026

Source label: Gemini 3.5 Flash (minimal)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 97%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

80.1%

#6 · kimi-k2.6

AA · Jun 24, 2026

Source label: Kimi K2.6

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 96.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

79.4%

#7 · Gemini 3 Flash

AA · Jun 24, 2026

Source label: Gemini 3 Flash Preview (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 95.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

78.6%

#8 · qwen3.6-plus

AA · Jun 24, 2026

Source label: Qwen3.6 Plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 94.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

78%

#9 · grok-build-0-1-06-16

AA · Jun 24, 2026

Source label: Grok Build 0.1 0616

verified runtimeexact direct

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 94.1%
Last updated: recent
Eligibility: headline eligible
Identity: exact (1.00)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

76.5%

#10 · Claude Opus 4.7

AA · Jun 24, 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: 93.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

76.4%

#11 · Gemini 3.1 Flash-Lite Preview

AA · Jun 24, 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: 92.6%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

75.5%

#12 · mimo-v2.5

AA · Jun 24, 2026

Source label: MiMo-V2.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 91.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

75.4%

#13 · Step 3.7 Flash

AA · Jun 24, 2026

Source label: Step 3.7 Flash

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 `mmmuPro`.

75.3%

#14 · Qwen3.6 35B A3B (Reasoning)

AA · Jun 24, 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: 90.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

75%

#15 · Gemini 2.5 Pro

AA · Jun 24, 2026

Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 89.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

74.9%

#16 · Qwen3.6 27B (Reasoning)

AA · Jun 24, 2026

Source label: Qwen3.6 27B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 88.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

74.6%

#17 · MiMo-V2-Omni-0327

AA · Jun 24, 2026

Source label: MiMo-V2-Omni-0327

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 88.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

73.9%

#18 · GPT-5.3 Codex

AA · Jun 24, 2026

Source label: GPT-5 Codex (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 87.4%
Last updated: recent
Eligibility: specialized_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

73.8%

#19 · Gemini 2.5 Flash Preview (Sep '25) (Reasoning)

AA · Jun 24, 2026

Source label: Gemini 2.5 Flash Preview (Sep '25) (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 86.7%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

73.1%

#20 · Kimi K2.5 (Reasoning)

AA · Jun 24, 2026

Source label: Kimi K2.5 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 85.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

73.1%

#21 · glm-5v-turbo

AA · Jun 24, 2026

Source label: GLM 5V Turbo (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 85.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

72.8%

#22 · Claude Opus 4.6

AA · Jun 24, 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: 84.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

72.5%

#23 · Qwen3.6 27B (Non-reasoning)

AA · Jun 24, 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: 83.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

71.7%

#24 · GPT-5.5

AA · Jun 24, 2026

Source label: GPT-5.5 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

71.4%

#25 · Claude Opus 4.5

AA · Jun 24, 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: 82.2%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

71.2%

#26 · Qwen3.6 35B A3B (Non-reasoning)

AA · Jun 24, 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: 81.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

71%

#27 · GPT-5.4

AA · Jun 24, 2026

Source label: GPT-5.4 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 80.7%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

70.6%

#28 · Qwen3.5 Omni Plus

AA · Jun 24, 2026

Source label: Qwen3.5 Omni Plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 80%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

70.5%

#29 · gemma-4-31b

AA · Jun 24, 2026

Source label: Gemma 4 31B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 79.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

70.3%

#30 · Qwen3.5 122B A10B

AA · Jun 24, 2026

Source label: Qwen3.5 122B A10B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 78.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

70.3%

#31 · o3

AA · Jun 24, 2026

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 77.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

70.1%

#32 · Qwen3.5 27B

AA · Jun 24, 2026

Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 77%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

70%

#33 · mimo-v2-omni

AA · Jun 24, 2026

Source label: MiMo-V2-Omni

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 `mmmuPro`.

69.9%

#34 · Gemma 4 12B (Reasoning)

AA · Jun 24, 2026

Source label: Gemma 4 12B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 75.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

69.7%

#35 · o4 mini

AA · Jun 24, 2026

Source label: o4-mini (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 74.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

69.2%

#36 · Claude Sonnet 4.6

AA · Jun 24, 2026

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

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 `mmmuPro`.

69.2%

#37 · Qwen3.5 35B A3B

AA · Jun 24, 2026

Source label: Qwen3.5 35B A3B (Non-reasoning)

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 `mmmuPro`.

69.2%

#38 · Grok 4

AA · Jun 24, 2026

Source label: Grok 4

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 72.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

68.8%

#39 · Qwen3 VL 235B A22B (Reasoning)

AA · Jun 24, 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: 71.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

68.7%

#40 · Doubao Seed Code

AA · Jun 24, 2026

Source label: Doubao Seed Code

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 71.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

68.1%

#41 · Claude Opus 4.1

AA · Jun 24, 2026

Source label: Claude 4.1 Opus (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 70.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

67.9%

#42 · Claude Opus 4

AA · Jun 24, 2026

Source label: Claude 4.1 Opus (Reasoning)

backfilledproxy backfilledBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 70.4%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.
Identity: benchmark proxy (0.58)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`. Backfilled from Claude Opus 4.1 via approved benchmark identity mapping map-claude-opus-4-to-4-1.

67.9%

#43 · qwen3-vl-235b-a22b-instruct

AA · Jun 24, 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: 68.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

67.6%

#44 · EXAONE 4.5 33B

AA · Jun 24, 2026

Source label: EXAONE 4.5 33B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 68.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

67.3%

#45 · Qwen3.5 9B

AA · Jun 24, 2026

Source label: Qwen3.5 9B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

66.8%

#46 · Gemma 4 26B A4B

AA · Jun 24, 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: 66.7%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

66.7%

#47 · GPT-5.2

AA · Jun 24, 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: 65.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

65.8%

#48 · Gemini 2.5 Flash

AA · Jun 24, 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: 65.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

65.5%

#49 · Qwen3.5 4B (Reasoning)

AA · Jun 24, 2026

Source label: Qwen3.5 4B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 64.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

65.4%

#50 · Claude Sonnet 4.5

AA · Jun 24, 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: 63.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

65.2%

#51 · Grok 4.20

AA · Jun 24, 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: 63%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64.9%

#52 · mistral-medium-3.5

AA · Jun 24, 2026

Source label: Mistral Medium 3.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 62.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64.9%

#53 · Grok 4.3

AA · Jun 24, 2026

Source label: Grok 4.3 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 61.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64.8%

#54 · Qwen3.5 Omni Flash

AA · Jun 24, 2026

Source label: Qwen3.5 Omni Flash

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 `mmmuPro`.

64.7%

#55 · ERNIE 5.0 Thinking Preview

AA · Jun 24, 2026

Source label: ERNIE 5.0 Thinking Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 60%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64.6%

#56 · Nova 2.0 Pro Preview (medium)

AA · Jun 24, 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: 59.3%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64.5%

#57 · Grok 4.20 0309 (Non-reasoning)

AA · Jun 24, 2026

Source label: Grok 4.20 0309 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64%

#58 · Step3 VL 10B

AA · Jun 24, 2026

Source label: Step3 VL 10B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 57.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64%

#59 · Nova 2.0 Lite (high)

AA · Jun 24, 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: 57%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

63.8%

#60 · Qwen3 VL 32B Instruct

AA · Jun 24, 2026

Source label: Qwen3 VL 32B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

63.4%

#61 · Command A+

AA · Jun 24, 2026

Source label: Command A+

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 55.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

63.2%

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

AA · Jun 24, 2026

Source label: Nova 2.0 Pro Preview (low)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 54.8%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.7%

#63 · Nova 2.0 Lite (medium)

AA · Jun 24, 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: 54.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.5%

#64 · GPT-5.1

AA · Jun 24, 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: 53.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.4%

#65 · Claude Sonnet 4

AA · Jun 24, 2026

Source label: Claude 4 Sonnet (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 53.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.4%

#66 · Llama 4 Maverick

AA · Jun 24, 2026

Source label: Llama 4 Maverick

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 51.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.1%

#67 · GPT-5

AA · Jun 24, 2026

Source label: GPT-5 (minimal)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 51.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.1%

#68 · Qwen3.5 4B (Non-reasoning)

AA · Jun 24, 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: 51.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.1%

#69 · Gemma 4 12B (Non-reasoning)

AA · Jun 24, 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: 49.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62%

#70 · Gemini 2.5 Flash Preview (Non-reasoning)

AA · Jun 24, 2026

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

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62%

#71 · Nova 2.0 Omni (medium)

AA · Jun 24, 2026

Source label: Nova 2.0 Omni (medium)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 48.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

61.9%

#72 · Qwen3 VL 30B A3B Instruct

AA · Jun 24, 2026

Source label: Qwen3 VL 30B A3B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 47.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

61.8%

#73 · Claude 4 Sonnet (Reasoning)

AA · Jun 24, 2026

Source label: Claude 4 Sonnet (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 46.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

61.8%

#74 · GPT-4.1

AA · Jun 24, 2026

Source label: GPT-4.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 45.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

61.2%

#75 · Qwen3 Omni 30B A3B (Reasoning)

AA · Jun 24, 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: 45.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

60.2%

#76 · Claude Sonnet 3.7

AA · Jun 24, 2026

Source label: Claude 3.7 Sonnet (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 44.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

60.1%

#77 · Nova 2.0 Omni (low)

AA · Jun 24, 2026

Source label: Nova 2.0 Omni (low)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 43.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

59.8%

#78 · Magistral Medium 1.2

AA · Jun 24, 2026

Source label: Magistral Medium 1.2

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 43%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

59.7%

#79 · GPT-4.1 mini

AA · Jun 24, 2026

Source label: GPT-4.1 mini

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 42.2%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

58.7%

#80 · Claude 4.5 Haiku (Reasoning)

AA · Jun 24, 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: 41.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

58.6%

#81 · GPT-5.4 mini

AA · Jun 24, 2026

Source label: GPT-5 mini (minimal)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 40.7%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

58.4%

#82 · Nova 2.0 Lite (low)

AA · Jun 24, 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: 40%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

58%

#83 · Qwen3 VL 8B (Reasoning)

AA · Jun 24, 2026

Source label: Qwen3 VL 8B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

56.6%

#84 · GPT-4o

AA · Jun 24, 2026

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

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 38.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

56.3%

#85 · Mistral Large 3

AA · Jun 24, 2026

Source label: Mistral Large 3

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 37.8%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

55.7%

#86 · Magistral Small 1.2

AA · Jun 24, 2026

Source label: Magistral Small 1.2

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 37%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

55.5%

#87 · Qwen3 Omni 30B A3B Instruct

AA · Jun 24, 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: 37%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

55.5%

#88 · Claude Haiku 4.5

AA · Jun 24, 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: 35.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

55.1%

#89 · Gemini 1.5 Pro

AA · Jun 24, 2026

Source label: Gemini 1.5 Pro (Sep '24)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

55%

#90 · Mistral Medium 3.1

AA · Jun 24, 2026

Source label: Mistral Medium 3.1

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 34.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

54.2%

#91 · Gemini 2.5 Flash-Lite

AA · Jun 24, 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: 33.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

54%

#92 · Nemotron 3 Nano Omni 30B A3B Reasoning

AA · Jun 24, 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: 32.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

53.2%

#93 · Mistral Medium 3

AA · Jun 24, 2026

Source label: Mistral Medium 3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 31.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

53%

#94 · Llama 4 Scout

AA · Jun 24, 2026

Source label: Llama 4 Scout

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 31.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

52.9%

#95 · Qwen3.5 397B A17B

AA · Jun 24, 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: 30.4%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

52.7%

#96 · Qwen3 VL 4B (Reasoning)

AA · Jun 24, 2026

Source label: Qwen3 VL 4B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 29.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

52%

#97 · Gemma 4 E4B (Reasoning)

AA · Jun 24, 2026

Source label: Gemma 4 E4B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 28.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

51.4%

#98 · Gemma 4 E4B (Non-reasoning)

AA · Jun 24, 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: 28.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

51.2%

#99 · pixtral-large-2411

AA · Jun 24, 2026

Source label: Pixtral Large

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 27.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

50.6%

#100 · Nova 2.0 Omni (Non-reasoning)

AA · Jun 24, 2026

Source label: Nova 2.0 Omni (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 26.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

49.9%

#101 · Ministral 3 14B

AA · Jun 24, 2026

Source label: Ministral 3 14B

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

49.8%

#102 · Nova 2.0 Lite (Non-reasoning)

AA · Jun 24, 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: 25.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

49%

#103 · Grok 4.1 Fast

AA · Jun 24, 2026

Source label: Grok 4.1 Fast (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 24.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

48.4%

#104 · Gemini 1.5 Flash

AA · Jun 24, 2026

Source label: Gemini 1.5 Flash (Sep '24)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

48.4%

#105 · Grok 4 Fast

AA · Jun 24, 2026

Source label: Grok 4 Fast (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 23%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

48.1%

#106 · Gemma 3 27B Instruct

AA · Jun 24, 2026

Source label: Gemma 3 27B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 22.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

48%

#107 · Mistral Small 3.2

AA · Jun 24, 2026

Source label: Mistral Small 3.2

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 21.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

48%

#108 · Qwen3 VL 8B Instruct

AA · Jun 24, 2026

Source label: Qwen3 VL 8B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 20.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

47.3%

#109 · Mistral Small 4

AA · Jun 24, 2026

Source label: Mistral Small 4 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 20%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

46.5%

#110 · Ministral 3 8B

AA · Jun 24, 2026

Source label: Ministral 3 8B

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 19.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

46%

#111 · Claude Haiku 3.5

AA · Jun 24, 2026

Source label: Claude 3.5 Haiku

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 18.5%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

45.6%

#112 · Devstral Small 2

AA · Jun 24, 2026

Source label: Devstral Small 2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 17.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

44.6%

#113 · Gemma 4 E2B (Reasoning)

AA · Jun 24, 2026

Source label: Gemma 4 E2B (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 `mmmuPro`.

44.6%

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

AA · Jun 24, 2026

Source label: NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 16.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

44.5%

#115 · Nova Pro

AA · Jun 24, 2026

Source label: Nova Pro

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 `mmmuPro`.

44.3%

#116 · Qwen3 VL 4B Instruct

AA · Jun 24, 2026

Source label: Qwen3 VL 4B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 14.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

43.9%

#117 · glm-4.5v

AA · Jun 24, 2026

Source label: GLM-4.5V (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 14.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

42.8%

#118 · Qwen3.5 2B (Non-reasoning)

AA · Jun 24, 2026

Source label: Qwen3.5 2B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 13.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

42.7%

#119 · glm-4.6v

AA · Jun 24, 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: 12.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

42.2%

#120 · Gemma 4 E2B (Non-reasoning)

AA · Jun 24, 2026

Source label: Gemma 4 E2B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 11.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

41.8%

#121 · GPT-4o mini

AA · Jun 24, 2026

Source label: GPT-4o mini

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 11.1%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

41.5%

#122 · GPT-4.1 nano

AA · Jun 24, 2026

Source label: GPT-4.1 nano

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 10.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

40.1%

#123 · Llama 3.2 Instruct 90B (Vision)

AA · Jun 24, 2026

Source label: Llama 3.2 Instruct 90B (Vision)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 9.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

39.5%

#124 · Ministral 3 3B

AA · Jun 24, 2026

Source label: Ministral 3 3B

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 `mmmuPro`.

38.1%

#125 · MiniCPM-V 4.6 1.3B

AA · Jun 24, 2026

Source label: MiniCPM-V 4.6 1.3B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 8.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

37.9%

#126 · Nova Lite

AA · Jun 24, 2026

Source label: Nova Lite

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 7.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

37.8%

#127 · Gemma 3 12B Instruct

AA · Jun 24, 2026

Source label: Gemma 3 12B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

37.5%

#128 · Gemini 1.5 Flash 8B

AA · Jun 24, 2026

Source label: Gemini 1.5 Flash-8B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 5.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

36.5%

#129 · GPT-5.4 nano

AA · Jun 24, 2026

Source label: GPT-5 nano (minimal)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 5.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

31.8%

#130 · Claude Haiku 3

AA · Jun 24, 2026

Source label: Claude 3 Haiku

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 4.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

30.8%

#131 · Gemma 3 4B Instruct

AA · Jun 24, 2026

Source label: Gemma 3 4B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 3.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

29.9%

#132 · Llama 3.2 Instruct 11B (Vision)

AA · Jun 24, 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: 3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

29.3%

#133 · LFM2.5-VL-1.6B

AA · Jun 24, 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: 2.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

26.5%

#134 · Gemma 3n E4B Instruct

AA · Jun 24, 2026

Source label: Gemma 3n E4B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 1.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

26.2%

#135 · Qwen3.5 0.8B (Non-reasoning)

AA · Jun 24, 2026

Source label: Qwen3.5 0.8B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 0.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

25.7%

#136 · Phi-4 Multimodal Instruct

AA · Jun 24, 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%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

14.5%

Benchmarks · /benchmarks/artificial-analysis-mmmu-pro

MMMU-Pro

MMMU-Pro result as reported through Artificial Analysis.

Source · Artificial Analysis
Version · artificial-analysis snapshot 2026-06-24
Scores · 136

Test details

Visible tradeoffsThis is an objective signal, so it is mainly about measurable task performance rather than public taste.

source

Artificial Analysis

metric

Score (%)

judge

Objective

direction

higher better

group id

aa_mmmu_pro_current

domain

Vision understanding

What it measures vs what it misses

✓ Measures

Multimodal reasoning over images and prompts.

✗ Misses

Adjacent capabilities, subjective preference, latency, and cost.

Leaderboard · this benchmark version

#1 · Gemini 3.1 Pro Preview

AA · Jun 24, 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: 100%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

82.4%

#2 · muse-spark

AA · Jun 24, 2026

Source label: Muse Spark

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 99.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

80.5%

#3 · Qwen3.7 Plus

AA · Jun 24, 2026

Source label: Qwen3.7 Plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 98.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

80.5%

#4 · Gemini 3 Pro Preview

AA · Jun 24, 2026

Source label: Gemini 3 Pro Preview (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

80.2%

#5 · Gemini 3.5 Flash

AA · Jun 24, 2026

Source label: Gemini 3.5 Flash (minimal)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 97%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

80.1%

#6 · kimi-k2.6

AA · Jun 24, 2026

Source label: Kimi K2.6

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 96.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

79.4%

#7 · Gemini 3 Flash

AA · Jun 24, 2026

Source label: Gemini 3 Flash Preview (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 95.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

78.6%

#8 · qwen3.6-plus

AA · Jun 24, 2026

Source label: Qwen3.6 Plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 94.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

78%

#9 · grok-build-0-1-06-16

AA · Jun 24, 2026

Source label: Grok Build 0.1 0616

verified runtimeexact direct

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 94.1%
Last updated: recent
Eligibility: headline eligible
Identity: exact (1.00)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

76.5%

#10 · Claude Opus 4.7

AA · Jun 24, 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: 93.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

76.4%

#11 · Gemini 3.1 Flash-Lite Preview

AA · Jun 24, 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: 92.6%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

75.5%

#12 · mimo-v2.5

AA · Jun 24, 2026

Source label: MiMo-V2.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 91.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

75.4%

#13 · Step 3.7 Flash

AA · Jun 24, 2026

Source label: Step 3.7 Flash

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 `mmmuPro`.

75.3%

#14 · Qwen3.6 35B A3B (Reasoning)

AA · Jun 24, 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: 90.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

75%

#15 · Gemini 2.5 Pro

AA · Jun 24, 2026

Source label: Gemini 2.5 Pro

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 89.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

74.9%

#16 · Qwen3.6 27B (Reasoning)

AA · Jun 24, 2026

Source label: Qwen3.6 27B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 88.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

74.6%

#17 · MiMo-V2-Omni-0327

AA · Jun 24, 2026

Source label: MiMo-V2-Omni-0327

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 88.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

73.9%

#18 · GPT-5.3 Codex

AA · Jun 24, 2026

Source label: GPT-5 Codex (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 87.4%
Last updated: recent
Eligibility: specialized_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

73.8%

#19 · Gemini 2.5 Flash Preview (Sep '25) (Reasoning)

AA · Jun 24, 2026

Source label: Gemini 2.5 Flash Preview (Sep '25) (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 86.7%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

73.1%

#20 · Kimi K2.5 (Reasoning)

AA · Jun 24, 2026

Source label: Kimi K2.5 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 85.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

73.1%

#21 · glm-5v-turbo

AA · Jun 24, 2026

Source label: GLM 5V Turbo (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 85.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

72.8%

#22 · Claude Opus 4.6

AA · Jun 24, 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: 84.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

72.5%

#23 · Qwen3.6 27B (Non-reasoning)

AA · Jun 24, 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: 83.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

71.7%

#24 · GPT-5.5

AA · Jun 24, 2026

Source label: GPT-5.5 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

71.4%

#25 · Claude Opus 4.5

AA · Jun 24, 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: 82.2%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

71.2%

#26 · Qwen3.6 35B A3B (Non-reasoning)

AA · Jun 24, 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: 81.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

71%

#27 · GPT-5.4

AA · Jun 24, 2026

Source label: GPT-5.4 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 80.7%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

70.6%

#28 · Qwen3.5 Omni Plus

AA · Jun 24, 2026

Source label: Qwen3.5 Omni Plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 80%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

70.5%

#29 · gemma-4-31b

AA · Jun 24, 2026

Source label: Gemma 4 31B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 79.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

70.3%

#30 · Qwen3.5 122B A10B

AA · Jun 24, 2026

Source label: Qwen3.5 122B A10B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 78.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

70.3%

#31 · o3

AA · Jun 24, 2026

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 77.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

70.1%

#32 · Qwen3.5 27B

AA · Jun 24, 2026

Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 77%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

70%

#33 · mimo-v2-omni

AA · Jun 24, 2026

Source label: MiMo-V2-Omni

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 `mmmuPro`.

69.9%

#34 · Gemma 4 12B (Reasoning)

AA · Jun 24, 2026

Source label: Gemma 4 12B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 75.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

69.7%

#35 · o4 mini

AA · Jun 24, 2026

Source label: o4-mini (high)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 74.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

69.2%

#36 · Claude Sonnet 4.6

AA · Jun 24, 2026

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

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 `mmmuPro`.

69.2%

#37 · Qwen3.5 35B A3B

AA · Jun 24, 2026

Source label: Qwen3.5 35B A3B (Non-reasoning)

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 `mmmuPro`.

69.2%

#38 · Grok 4

AA · Jun 24, 2026

Source label: Grok 4

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 72.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

68.8%

#39 · Qwen3 VL 235B A22B (Reasoning)

AA · Jun 24, 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: 71.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

68.7%

#40 · Doubao Seed Code

AA · Jun 24, 2026

Source label: Doubao Seed Code

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 71.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

68.1%

#41 · Claude Opus 4.1

AA · Jun 24, 2026

Source label: Claude 4.1 Opus (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 70.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

67.9%

#42 · Claude Opus 4

AA · Jun 24, 2026

Source label: Claude 4.1 Opus (Reasoning)

backfilledproxy backfilledBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 70.4%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.
Identity: benchmark proxy (0.58)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`. Backfilled from Claude Opus 4.1 via approved benchmark identity mapping map-claude-opus-4-to-4-1.

67.9%

#43 · qwen3-vl-235b-a22b-instruct

AA · Jun 24, 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: 68.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

67.6%

#44 · EXAONE 4.5 33B

AA · Jun 24, 2026

Source label: EXAONE 4.5 33B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 68.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

67.3%

#45 · Qwen3.5 9B

AA · Jun 24, 2026

Source label: Qwen3.5 9B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

66.8%

#46 · Gemma 4 26B A4B

AA · Jun 24, 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: 66.7%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

66.7%

#47 · GPT-5.2

AA · Jun 24, 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: 65.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

65.8%

#48 · Gemini 2.5 Flash

AA · Jun 24, 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: 65.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

65.5%

#49 · Qwen3.5 4B (Reasoning)

AA · Jun 24, 2026

Source label: Qwen3.5 4B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 64.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

65.4%

#50 · Claude Sonnet 4.5

AA · Jun 24, 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: 63.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

65.2%

#51 · Grok 4.20

AA · Jun 24, 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: 63%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64.9%

#52 · mistral-medium-3.5

AA · Jun 24, 2026

Source label: Mistral Medium 3.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 62.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64.9%

#53 · Grok 4.3

AA · Jun 24, 2026

Source label: Grok 4.3 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 61.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64.8%

#54 · Qwen3.5 Omni Flash

AA · Jun 24, 2026

Source label: Qwen3.5 Omni Flash

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 `mmmuPro`.

64.7%

#55 · ERNIE 5.0 Thinking Preview

AA · Jun 24, 2026

Source label: ERNIE 5.0 Thinking Preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 60%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64.6%

#56 · Nova 2.0 Pro Preview (medium)

AA · Jun 24, 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: 59.3%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64.5%

#57 · Grok 4.20 0309 (Non-reasoning)

AA · Jun 24, 2026

Source label: Grok 4.20 0309 (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64%

#58 · Step3 VL 10B

AA · Jun 24, 2026

Source label: Step3 VL 10B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 57.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

64%

#59 · Nova 2.0 Lite (high)

AA · Jun 24, 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: 57%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

63.8%

#60 · Qwen3 VL 32B Instruct

AA · Jun 24, 2026

Source label: Qwen3 VL 32B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

63.4%

#61 · Command A+

AA · Jun 24, 2026

Source label: Command A+

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 55.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

63.2%

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

AA · Jun 24, 2026

Source label: Nova 2.0 Pro Preview (low)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 54.8%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.7%

#63 · Nova 2.0 Lite (medium)

AA · Jun 24, 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: 54.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.5%

#64 · GPT-5.1

AA · Jun 24, 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: 53.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.4%

#65 · Claude Sonnet 4

AA · Jun 24, 2026

Source label: Claude 4 Sonnet (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 53.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.4%

#66 · Llama 4 Maverick

AA · Jun 24, 2026

Source label: Llama 4 Maverick

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 51.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.1%

#67 · GPT-5

AA · Jun 24, 2026

Source label: GPT-5 (minimal)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 51.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.1%

#68 · Qwen3.5 4B (Non-reasoning)

AA · Jun 24, 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: 51.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62.1%

#69 · Gemma 4 12B (Non-reasoning)

AA · Jun 24, 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: 49.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62%

#70 · Gemini 2.5 Flash Preview (Non-reasoning)

AA · Jun 24, 2026

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

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

62%

#71 · Nova 2.0 Omni (medium)

AA · Jun 24, 2026

Source label: Nova 2.0 Omni (medium)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 48.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

61.9%

#72 · Qwen3 VL 30B A3B Instruct

AA · Jun 24, 2026

Source label: Qwen3 VL 30B A3B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 47.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

61.8%

#73 · Claude 4 Sonnet (Reasoning)

AA · Jun 24, 2026

Source label: Claude 4 Sonnet (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 46.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

61.8%

#74 · GPT-4.1

AA · Jun 24, 2026

Source label: GPT-4.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 45.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

61.2%

#75 · Qwen3 Omni 30B A3B (Reasoning)

AA · Jun 24, 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: 45.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

60.2%

#76 · Claude Sonnet 3.7

AA · Jun 24, 2026

Source label: Claude 3.7 Sonnet (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 44.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

60.1%

#77 · Nova 2.0 Omni (low)

AA · Jun 24, 2026

Source label: Nova 2.0 Omni (low)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 43.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

59.8%

#78 · Magistral Medium 1.2

AA · Jun 24, 2026

Source label: Magistral Medium 1.2

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 43%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

59.7%

#79 · GPT-4.1 mini

AA · Jun 24, 2026

Source label: GPT-4.1 mini

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 42.2%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

58.7%

#80 · Claude 4.5 Haiku (Reasoning)

AA · Jun 24, 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: 41.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

58.6%

#81 · GPT-5.4 mini

AA · Jun 24, 2026

Source label: GPT-5 mini (minimal)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 40.7%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

58.4%

#82 · Nova 2.0 Lite (low)

AA · Jun 24, 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: 40%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

58%

#83 · Qwen3 VL 8B (Reasoning)

AA · Jun 24, 2026

Source label: Qwen3 VL 8B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

56.6%

#84 · GPT-4o

AA · Jun 24, 2026

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

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 38.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

56.3%

#85 · Mistral Large 3

AA · Jun 24, 2026

Source label: Mistral Large 3

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 37.8%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

55.7%

#86 · Magistral Small 1.2

AA · Jun 24, 2026

Source label: Magistral Small 1.2

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 37%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

55.5%

#87 · Qwen3 Omni 30B A3B Instruct

AA · Jun 24, 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: 37%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

55.5%

#88 · Claude Haiku 4.5

AA · Jun 24, 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: 35.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

55.1%

#89 · Gemini 1.5 Pro

AA · Jun 24, 2026

Source label: Gemini 1.5 Pro (Sep '24)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

55%

#90 · Mistral Medium 3.1

AA · Jun 24, 2026

Source label: Mistral Medium 3.1

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 34.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

54.2%

#91 · Gemini 2.5 Flash-Lite

AA · Jun 24, 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: 33.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

54%

#92 · Nemotron 3 Nano Omni 30B A3B Reasoning

AA · Jun 24, 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: 32.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

53.2%

#93 · Mistral Medium 3

AA · Jun 24, 2026

Source label: Mistral Medium 3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 31.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

53%

#94 · Llama 4 Scout

AA · Jun 24, 2026

Source label: Llama 4 Scout

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 31.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

52.9%

#95 · Qwen3.5 397B A17B

AA · Jun 24, 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: 30.4%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

52.7%

#96 · Qwen3 VL 4B (Reasoning)

AA · Jun 24, 2026

Source label: Qwen3 VL 4B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 29.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

52%

#97 · Gemma 4 E4B (Reasoning)

AA · Jun 24, 2026

Source label: Gemma 4 E4B (Reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 28.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

51.4%

#98 · Gemma 4 E4B (Non-reasoning)

AA · Jun 24, 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: 28.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

51.2%

#99 · pixtral-large-2411

AA · Jun 24, 2026

Source label: Pixtral Large

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 27.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

50.6%

#100 · Nova 2.0 Omni (Non-reasoning)

AA · Jun 24, 2026

Source label: Nova 2.0 Omni (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 26.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

49.9%

#101 · Ministral 3 14B

AA · Jun 24, 2026

Source label: Ministral 3 14B

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

49.8%

#102 · Nova 2.0 Lite (Non-reasoning)

AA · Jun 24, 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: 25.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

49%

#103 · Grok 4.1 Fast

AA · Jun 24, 2026

Source label: Grok 4.1 Fast (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 24.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

48.4%

#104 · Gemini 1.5 Flash

AA · Jun 24, 2026

Source label: Gemini 1.5 Flash (Sep '24)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

48.4%

#105 · Grok 4 Fast

AA · Jun 24, 2026

Source label: Grok 4 Fast (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 23%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

48.1%

#106 · Gemma 3 27B Instruct

AA · Jun 24, 2026

Source label: Gemma 3 27B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 22.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

48%

#107 · Mistral Small 3.2

AA · Jun 24, 2026

Source label: Mistral Small 3.2

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 21.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

48%

#108 · Qwen3 VL 8B Instruct

AA · Jun 24, 2026

Source label: Qwen3 VL 8B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 20.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

47.3%

#109 · Mistral Small 4

AA · Jun 24, 2026

Source label: Mistral Small 4 (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 20%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

46.5%

#110 · Ministral 3 8B

AA · Jun 24, 2026

Source label: Ministral 3 8B

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 19.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

46%

#111 · Claude Haiku 3.5

AA · Jun 24, 2026

Source label: Claude 3.5 Haiku

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 18.5%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

45.6%

#112 · Devstral Small 2

AA · Jun 24, 2026

Source label: Devstral Small 2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 17.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

44.6%

#113 · Gemma 4 E2B (Reasoning)

AA · Jun 24, 2026

Source label: Gemma 4 E2B (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 `mmmuPro`.

44.6%

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

AA · Jun 24, 2026

Source label: NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 16.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

44.5%

#115 · Nova Pro

AA · Jun 24, 2026

Source label: Nova Pro

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 `mmmuPro`.

44.3%

#116 · Qwen3 VL 4B Instruct

AA · Jun 24, 2026

Source label: Qwen3 VL 4B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 14.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

43.9%

#117 · glm-4.5v

AA · Jun 24, 2026

Source label: GLM-4.5V (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 14.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

42.8%

#118 · Qwen3.5 2B (Non-reasoning)

AA · Jun 24, 2026

Source label: Qwen3.5 2B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 13.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

42.7%

#119 · glm-4.6v

AA · Jun 24, 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: 12.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

42.2%

#120 · Gemma 4 E2B (Non-reasoning)

AA · Jun 24, 2026

Source label: Gemma 4 E2B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 11.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

41.8%

#121 · GPT-4o mini

AA · Jun 24, 2026

Source label: GPT-4o mini

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 11.1%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

41.5%

#122 · GPT-4.1 nano

AA · Jun 24, 2026

Source label: GPT-4.1 nano

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 10.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

40.1%

#123 · Llama 3.2 Instruct 90B (Vision)

AA · Jun 24, 2026

Source label: Llama 3.2 Instruct 90B (Vision)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 9.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

39.5%

#124 · Ministral 3 3B

AA · Jun 24, 2026

Source label: Ministral 3 3B

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 `mmmuPro`.

38.1%

#125 · MiniCPM-V 4.6 1.3B

AA · Jun 24, 2026

Source label: MiniCPM-V 4.6 1.3B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 8.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

37.9%

#126 · Nova Lite

AA · Jun 24, 2026

Source label: Nova Lite

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 7.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

37.8%

#127 · Gemma 3 12B Instruct

AA · Jun 24, 2026

Source label: Gemma 3 12B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

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

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

37.5%

#128 · Gemini 1.5 Flash 8B

AA · Jun 24, 2026

Source label: Gemini 1.5 Flash-8B

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 5.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

36.5%

#129 · GPT-5.4 nano

AA · Jun 24, 2026

Source label: GPT-5 nano (minimal)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 5.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

31.8%

#130 · Claude Haiku 3

AA · Jun 24, 2026

Source label: Claude 3 Haiku

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 4.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

30.8%

#131 · Gemma 3 4B Instruct

AA · Jun 24, 2026

Source label: Gemma 3 4B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 3.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

29.9%

#132 · Llama 3.2 Instruct 11B (Vision)

AA · Jun 24, 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: 3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

29.3%

#133 · LFM2.5-VL-1.6B

AA · Jun 24, 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: 2.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

26.5%

#134 · Gemma 3n E4B Instruct

AA · Jun 24, 2026

Source label: Gemma 3n E4B Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 1.5%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

26.2%

#135 · Qwen3.5 0.8B (Non-reasoning)

AA · Jun 24, 2026

Source label: Qwen3.5 0.8B (Non-reasoning)

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://artificialanalysis.ai/leaderboards/models
Percentile: 0.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

25.7%

#136 · Phi-4 Multimodal Instruct

AA · Jun 24, 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%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.94)

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

14.5%

MMMU-Pro

Test details

What it measures vs what it misses

✓ Measures

✗ Misses

Leaderboard · this benchmark version

Loading benchmark evidence.

MMMU-Pro

Test details

What it measures vs what it misses

✓ Measures

✗ Misses

Leaderboard · this benchmark version