Benchmarks · /benchmarks/vals-ai-legal_bench

LegalBench

Name: LegalBench
Creator: Vals AI

LegalBench result as reported by Vals AI.

Source · Vals AI
Version · vals-ai snapshot 2026-06-24
Scores · 91

Test details

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

source

Vals AI

metric

Accuracy (%)

judge

Objective

direction

higher better

group id

vals_legal_bench_current

domain

Professional reasoning

What it measures vs what it misses

✓ Measures

Academic legal reasoning tasks.

✗ Misses

Adjacent skills outside the benchmark task mix, latency, and cost.

Why this countsAcademic legal reasoning tasks.Same-test ruleThis percentile only compares models inside the exact benchmark/version group shown here. It is not a universal score.What it missesAdjacent skills outside the benchmark task mix, latency, and cost.

Leaderboard · this benchmark version

#1 · Claude Fable 5

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-fable-5

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 100%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

88.6%87.9% - 89.2%

#2 · Gemini 3.1 Pro Preview

VALS-AI · Jun 17, 2026

Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 98.9%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

87.4%86.8% - 88%

#3 · Gemini 3 Pro Preview

VALS-AI · Jun 17, 2026

Source label: google/gemini-3-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 97.8%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

87%86.3% - 87.7%

#4 · Gemini 3 Flash

VALS-AI · Jun 17, 2026

Source label: google/gemini-3-flash-preview

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 96.7%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

86.9%86.2% - 87.6%

#5 · GPT-5.5

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.5

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 95.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

86.5%85.7% - 87.3%

#6 · GPT-5.4

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.4-2026-03-05

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 94.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

86%85.2% - 86.9%

#7 · GPT-5

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5-2025-08-07

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 93.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

86%85.3% - 86.8%

#8 · GPT-5.1

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.1-2025-11-13

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 92.2%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

85.7%84.9% - 86.5%

#9 · minimax-m3

VALS-AI · Jun 17, 2026

Source label: minimax/MiniMax-M3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 91.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: MiniMax.

85.4%84.6% - 86.3%

#10 · Claude Opus 4.6

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-6-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 90%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

85.3%84.6% - 86%

#11 · Claude Opus 4.7

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-7

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 88.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

85.3%84.4% - 86.1%

#12 · Qwen3.5 Plus

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3.5-plus-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 87.8%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Alibaba.

85.1%84.3% - 85.9%

#13 · Qwen3.7 Max

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3.7-max

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 86.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Alibaba.

84.9%84.1% - 85.7%

#14 · kimi-k2.6

VALS-AI · Jun 17, 2026

Source label: kimi/kimi-k2.6

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 85.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Moonshot AI.

84.7%83.9% - 85.6%

#15 · Claude Opus 4.5

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-5-20251101-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 84.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

84.6%83.8% - 85.4%

#16 · Grok 4.3

VALS-AI · Jun 17, 2026

Source label: grok/grok-4.3

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 83.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

84.5%83.6% - 85.3%

#17 · glm-5.1

VALS-AI · Jun 17, 2026

Source label: zai/glm-5.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 82.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

84.4%83.6% - 85.2%

#18 · Gemini 2.5 Pro

VALS-AI · Jun 17, 2026

Source label: google/gemini-2.5-pro-exp-03-25

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 81.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

84.3%83.5% - 85.1%

#19 · Qwen3.5 Flash

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3.5-flash

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 80%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Alibaba.

84.3%83.5% - 85.1%

#20 · qwen3.6-plus

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3.6-plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 78.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Alibaba.

84.2%83.4% - 85.1%

#21 · muse-spark

VALS-AI · Jun 17, 2026

Source label: meta/muse_spark

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 77.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Meta.

84.2%83.4% - 85%

#22 · Claude Sonnet 4.5

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-sonnet-4-5-20250929-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 76.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

84.1%83.2% - 85%

#23 · GLM-5.2 (max)

VALS-AI · Jun 17, 2026

Source label: zai/glm-5.2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 75.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

84.1%83.2% - 85%

#24 · glm-5

VALS-AI · Jun 17, 2026

Source label: zai/glm-5-thinking

verified runtimeexact directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 74.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: exact (1.00)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

84.1%83.3% - 84.9%

#25 · minimax-m2.7

VALS-AI · Jun 17, 2026

Source label: minimax/MiniMax-M2.7

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 73.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: MiniMax.

84%83.2% - 84.8%

#26 · Gemini 2.5 Flash

VALS-AI · Jun 17, 2026

Source label: google/gemini-2.5-flash-preview-04-17

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 72.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

83.8%83% - 84.6%

#27 · Gemini 3.1 Flash-Lite Preview

VALS-AI · Jun 17, 2026

Source label: google/gemini-3.1-flash-lite-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 71.1%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

83.8%83% - 84.6%

#28 · o3

VALS-AI · Jun 17, 2026

Source label: openai/o3-2025-04-16

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 70%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

83.8%82.9% - 84.6%

#29 · Gemini 3.5 Flash

VALS-AI · Jun 17, 2026

Source label: google/gemini-3.5-flash

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 68.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

83.6%81.9% - 85.3%

#30 · Claude Opus 4.8

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-8

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 67.8%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

83.6%82.6% - 84.5%

#31 · Claude Opus 4.1

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-1-20250805

verified runtimevariant directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 66.7%
Last updated: recent
Eligibility: historical_model
Identity: dated variant (0.80)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

83.5%82.6% - 84.3%

#32 · glm-4.7

VALS-AI · Jun 17, 2026

Source label: zai/glm-4.7

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 65.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

83.4%82.4% - 84.3%

#33 · Grok 4

VALS-AI · Jun 17, 2026

Source label: grok/grok-4-0709

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 64.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

83.2%82.3% - 84.1%

#34 · Grok 3 mini

VALS-AI · Jun 17, 2026

Source label: grok/grok-3-mini-fast-high-reasoning

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 63.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

83.1%82.3% - 84%

#35 · GPT-4.1

VALS-AI · Jun 17, 2026

Source label: openai/gpt-4.1-2025-04-14

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 62.2%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

83.1%82.3% - 83.9%

#36 · Claude Opus 4

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-20250514

verified runtimevariant directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 61.1%
Last updated: recent
Eligibility: historical_model
Identity: dated variant (0.80)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

83.1%82.2% - 83.9%

#37 · Claude Sonnet 4

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-sonnet-4-20250514

verified runtimevariant directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 60%
Last updated: recent
Eligibility: historical_model
Identity: dated variant (0.80)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

83%82.2% - 83.7%

#38 · GPT-5.2

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.2-2025-12-11

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 58.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

82.8%82% - 83.5%

#39 · Grok 3

VALS-AI · Jun 17, 2026

Source label: grok/grok-3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 57.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

82.6%81.8% - 83.4%

#40 · Claude Sonnet 3.7

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-3-7-sonnet-20250219-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 56.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

82.5%81.5% - 83.6%

#41 · Grok 4.1 Fast

VALS-AI · Jun 17, 2026

Source label: grok/grok-4-1-fast-reasoning

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 55.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

82.5%81.6% - 83.3%

#42 · GPT-4o

VALS-AI · Jun 17, 2026

Source label: openai/gpt-4o-2024-11-20

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 54.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

82.2%81.3% - 83.1%

#43 · Claude Sonnet 4.6

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-sonnet-4-6

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 53.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

82.1%81.2% - 83.1%

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

VALS-AI · Jun 17, 2026

Source label: nvidia/nemotron-3-ultra-550b-a55b

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 52.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Nvidia.

82.1%81.2% - 83%

#45 · Gemini 2.5 Flash-Lite

VALS-AI · Jun 17, 2026

Source label: google/gemini-2.5-flash-lite-preview-09-2025-thinking

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 51.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

82%81.2% - 82.9%

#46 · Qwen3 Max

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3-max

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 50%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Alibaba.

81.9%81.1% - 82.6%

#47 · GPT-5.4 mini

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5-mini-2025-08-07

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 48.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

81.8%80.9% - 82.7%

#48 · kimi-k2-instruct

VALS-AI · Jun 17, 2026

Source label: together/moonshotai/Kimi-K2-Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 47.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Together AI.

81.5%80.6% - 82.3%

#49 · Claude Haiku 4.5

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-haiku-4-5-20251001-thinking

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 46.7%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

81.2%80.3% - 82.2%

#50 · deepseek-v3

VALS-AI · Jun 17, 2026

Source label: fireworks/deepseek-v3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 45.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Fireworks AI.

80.8%79.9% - 81.6%

#51 · Grok 4 Fast

VALS-AI · Jun 17, 2026

Source label: grok/grok-4-fast-reasoning

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 44.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

80.6%79.7% - 81.5%

#52 · GPT-4 Turbo

VALS-AI · Jun 17, 2026

Source label: openai/gpt-4-turbo

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 43.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

80.5%79.5% - 81.4%

#53 · o1

VALS-AI · Jun 17, 2026

Source label: openai/o1-2024-12-17

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 42.2%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

80.4%79.3% - 81.5%

#54 · deepseek-v4-pro

VALS-AI · Jun 17, 2026

Source label: deepseek/deepseek-v4-pro

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 41.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: DeepSeek.

80.3%79.4% - 81.2%

#55 · Kimi K2 Thinking

VALS-AI · Jun 17, 2026

Source label: kimi/kimi-k2-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 40%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Moonshot AI.

80.2%79.3% - 81.1%

#56 · Qwen3 235B A22B

VALS-AI · Jun 17, 2026

Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 38.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Fireworks AI.

80.2%79.3% - 81%

#57 · minimax-m2.5

VALS-AI · Jun 17, 2026

Source label: minimax/MiniMax-M2.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 37.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: MiniMax.

80%79% - 80.9%

#58 · command-a-03-2025

VALS-AI · Jun 17, 2026

Source label: cohere/command-a-03-2025

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 36.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Cohere.

79.7%78.9% - 80.5%

#59 · glm-4.6

VALS-AI · Jun 17, 2026

Source label: zai/glm-4.6

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 35.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

79.6%78.6% - 80.6%

#60 · o4 mini

VALS-AI · Jun 17, 2026

Source label: openai/o4-mini-2025-04-16

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 34.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

79.2%78.2% - 80.2%

#61 · Mistral Large (Feb '24)

VALS-AI · Jun 17, 2026

Source label: mistralai/mistral-large-2512

verified runtimevariant directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 33.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: dated variant (0.80)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Mistral AI.

79.1%78.2% - 80%

#62 · mimo-v2.5

VALS-AI · Jun 17, 2026

Source label: xiaomi/mimo-v2.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 32.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Xiaomi.

78.9%77.9% - 79.9%

#63 · Gemini 2.0 Flash

VALS-AI · Jun 17, 2026

Source label: google/gemini-2.0-flash-001

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 31.1%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

78.4%76.8% - 79.9%

#64 · GPT-4.1 mini

VALS-AI · Jun 17, 2026

Source label: openai/gpt-4.1-mini-2025-04-14

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 30%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

78%77.2% - 78.9%

#65 · GPT-5.4 nano

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 28.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

77.9%77.1% - 78.8%

#66 · Grok 4.20

VALS-AI · Jun 17, 2026

Source label: grok/grok-4.20-0309-reasoning

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 27.8%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

77.7%76.8% - 78.7%

#67 · deepseek-v3-0324

VALS-AI · Jun 17, 2026

Source label: fireworks/deepseek-v3-0324

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 26.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Fireworks AI.

77.7%76.9% - 78.5%

#68 · mimo-v2.5-pro

VALS-AI · Jun 17, 2026

Source label: xiaomi/mimo-v2.5-pro

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 25.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Xiaomi.

77.1%76.1% - 78.2%

#69 · GPT-OSS 120B

VALS-AI · Jun 17, 2026

Source label: fireworks/gpt-oss-120b

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 24.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Fireworks AI.

75.9%74.9% - 77%

#70 · glm-4.5

VALS-AI · Jun 17, 2026

Source label: zai/glm-4.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 23.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

75.6%74.6% - 76.6%

#71 · MiniMax-M2.1

VALS-AI · Jun 17, 2026

Source label: minimax/MiniMax-M2.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 22.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: MiniMax.

75.4%74.4% - 76.5%

#72 · jamba-1.5-large

VALS-AI · Jun 17, 2026

Source label: ai21labs/jamba-1.5-large

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 21.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: AI21 Labs.

74.2%72.9% - 75.4%

#73 · llama-4-scout-17b-16e-instruct

VALS-AI · Jun 17, 2026

Source label: together/meta-llama/Llama-4-Scout-17B-16E-Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 20%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Together AI.

72%71.1% - 73%

#74 · o3 mini

VALS-AI · Jun 17, 2026

Source label: openai/o3-mini-2025-01-31

verified runtimeexact directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 18.9%
Last updated: recent
Eligibility: historical_model
Identity: exact (1.00)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

71.5%70.1% - 73%

#75 · GPT-OSS 20B

VALS-AI · Jun 17, 2026

Source label: fireworks/gpt-oss-20b

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 17.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Fireworks AI.

70.8%69.7% - 72%

#76 · Claude Haiku 3.5

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-3-5-haiku-20241022

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 16.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

70.3%69.1% - 71.6%

#77 · Gemini 1.0 Pro

VALS-AI · Jun 17, 2026

Source label: google/gemini-1.0-pro-002

verified runtimevariant directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 15.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: dated variant (0.80)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

70.2%68.7% - 71.7%

#78 · Mistral Small (Sep '24)

VALS-AI · Jun 17, 2026

Source label: mistralai/mistral-small-2503

verified runtimevariant directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 14.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: dated variant (0.80)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Mistral AI.

69.2%68.2% - 70.2%

#79 · Gemini 1.5 Pro

VALS-AI · Jun 17, 2026

Source label: google/gemini-1.5-pro-002

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 13.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

69.1%67.6% - 70.6%

#80 · command-r-plus

VALS-AI · Jun 17, 2026

Source label: cohere/command-r-plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 12.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Cohere.

68.3%67% - 69.6%

#81 · gemma-2-9b-it

VALS-AI · Jun 17, 2026

Source label: together/google/gemma-2-9b-it

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 11.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Together AI.

68.1%66.7% - 69.6%

#82 · gemma-2-27b-it

VALS-AI · Jun 17, 2026

Source label: together/google/gemma-2-27b-it

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 10%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Together AI.

68%66.7% - 69.3%

#83 · DeepSeek Reasoner

VALS-AI · Jun 17, 2026

Source label: fireworks/deepseek-r1

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 8.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Fireworks AI.

67.3%66.3% - 68.4%

#84 · jamba-1.5-mini

VALS-AI · Jun 17, 2026

Source label: ai21labs/jamba-1.5-mini

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 7.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: AI21 Labs.

66.6%65.3% - 67.9%

#85 · GPT-3.5 Turbo

VALS-AI · Jun 17, 2026

Source label: openai/gpt-3.5-turbo

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 6.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

64.4%63.1% - 65.6%

#86 · Gemini 1.5 Flash

VALS-AI · Jun 17, 2026

Source label: google/gemini-1.5-flash-002

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 5.6%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

63.2%61.8% - 64.7%

#87 · mistral-medium-2505

VALS-AI · Jun 17, 2026

Source label: mistralai/mistral-medium-2505

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 4.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Mistral AI.

62%61.1% - 62.9%

#88 · GPT-4.1 nano

VALS-AI · Jun 17, 2026

Source label: openai/gpt-4.1-nano-2025-04-14

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 3.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

61.1%59.9% - 62.2%

#89 · Magistral Medium 1.2

VALS-AI · Jun 17, 2026

Source label: mistralai/magistral-medium-2509

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 2.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Mistral AI.

54.8%53.7% - 55.9%

#90 · Magistral Small 1.2

VALS-AI · Jun 17, 2026

Source label: mistralai/magistral-small-2509

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 1.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Mistral AI.

40%39.1% - 40.9%

#91 · command-r

VALS-AI · Jun 17, 2026

Source label: cohere/command-r

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 0%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Cohere.

33%31.6% - 34.4%

Benchmarks · /benchmarks/vals-ai-legal_bench

LegalBench

LegalBench result as reported by Vals AI.

Source · Vals AI
Version · vals-ai snapshot 2026-06-24
Scores · 91

Test details

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

source

Vals AI

metric

Accuracy (%)

judge

Objective

direction

higher better

group id

vals_legal_bench_current

domain

Professional reasoning

What it measures vs what it misses

✓ Measures

Academic legal reasoning tasks.

✗ Misses

Adjacent skills outside the benchmark task mix, latency, and cost.

Leaderboard · this benchmark version

#1 · Claude Fable 5

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-fable-5

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 100%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

88.6%87.9% - 89.2%

#2 · Gemini 3.1 Pro Preview

VALS-AI · Jun 17, 2026

Source label: google/gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 98.9%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

87.4%86.8% - 88%

#3 · Gemini 3 Pro Preview

VALS-AI · Jun 17, 2026

Source label: google/gemini-3-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 97.8%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

87%86.3% - 87.7%

#4 · Gemini 3 Flash

VALS-AI · Jun 17, 2026

Source label: google/gemini-3-flash-preview

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 96.7%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

86.9%86.2% - 87.6%

#5 · GPT-5.5

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.5

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 95.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

86.5%85.7% - 87.3%

#6 · GPT-5.4

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.4-2026-03-05

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 94.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

86%85.2% - 86.9%

#7 · GPT-5

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5-2025-08-07

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 93.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

86%85.3% - 86.8%

#8 · GPT-5.1

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.1-2025-11-13

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 92.2%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

85.7%84.9% - 86.5%

#9 · minimax-m3

VALS-AI · Jun 17, 2026

Source label: minimax/MiniMax-M3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 91.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: MiniMax.

85.4%84.6% - 86.3%

#10 · Claude Opus 4.6

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-6-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 90%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

85.3%84.6% - 86%

#11 · Claude Opus 4.7

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-7

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 88.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

85.3%84.4% - 86.1%

#12 · Qwen3.5 Plus

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3.5-plus-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 87.8%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Alibaba.

85.1%84.3% - 85.9%

#13 · Qwen3.7 Max

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3.7-max

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 86.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Alibaba.

84.9%84.1% - 85.7%

#14 · kimi-k2.6

VALS-AI · Jun 17, 2026

Source label: kimi/kimi-k2.6

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 85.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Moonshot AI.

84.7%83.9% - 85.6%

#15 · Claude Opus 4.5

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-5-20251101-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 84.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

84.6%83.8% - 85.4%

#16 · Grok 4.3

VALS-AI · Jun 17, 2026

Source label: grok/grok-4.3

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 83.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

84.5%83.6% - 85.3%

#17 · glm-5.1

VALS-AI · Jun 17, 2026

Source label: zai/glm-5.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 82.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

84.4%83.6% - 85.2%

#18 · Gemini 2.5 Pro

VALS-AI · Jun 17, 2026

Source label: google/gemini-2.5-pro-exp-03-25

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 81.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

84.3%83.5% - 85.1%

#19 · Qwen3.5 Flash

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3.5-flash

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 80%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Alibaba.

84.3%83.5% - 85.1%

#20 · qwen3.6-plus

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3.6-plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 78.9%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Alibaba.

84.2%83.4% - 85.1%

#21 · muse-spark

VALS-AI · Jun 17, 2026

Source label: meta/muse_spark

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 77.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Meta.

84.2%83.4% - 85%

#22 · Claude Sonnet 4.5

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-sonnet-4-5-20250929-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 76.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

84.1%83.2% - 85%

#23 · GLM-5.2 (max)

VALS-AI · Jun 17, 2026

Source label: zai/glm-5.2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 75.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

84.1%83.2% - 85%

#24 · glm-5

VALS-AI · Jun 17, 2026

Source label: zai/glm-5-thinking

verified runtimeexact directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 74.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: exact (1.00)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

84.1%83.3% - 84.9%

#25 · minimax-m2.7

VALS-AI · Jun 17, 2026

Source label: minimax/MiniMax-M2.7

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 73.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: MiniMax.

84%83.2% - 84.8%

#26 · Gemini 2.5 Flash

VALS-AI · Jun 17, 2026

Source label: google/gemini-2.5-flash-preview-04-17

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 72.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

83.8%83% - 84.6%

#27 · Gemini 3.1 Flash-Lite Preview

VALS-AI · Jun 17, 2026

Source label: google/gemini-3.1-flash-lite-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 71.1%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

83.8%83% - 84.6%

#28 · o3

VALS-AI · Jun 17, 2026

Source label: openai/o3-2025-04-16

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 70%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

83.8%82.9% - 84.6%

#29 · Gemini 3.5 Flash

VALS-AI · Jun 17, 2026

Source label: google/gemini-3.5-flash

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 68.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

83.6%81.9% - 85.3%

#30 · Claude Opus 4.8

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-8

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 67.8%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

83.6%82.6% - 84.5%

#31 · Claude Opus 4.1

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-1-20250805

verified runtimevariant directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 66.7%
Last updated: recent
Eligibility: historical_model
Identity: dated variant (0.80)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

83.5%82.6% - 84.3%

#32 · glm-4.7

VALS-AI · Jun 17, 2026

Source label: zai/glm-4.7

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 65.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

83.4%82.4% - 84.3%

#33 · Grok 4

VALS-AI · Jun 17, 2026

Source label: grok/grok-4-0709

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 64.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

83.2%82.3% - 84.1%

#34 · Grok 3 mini

VALS-AI · Jun 17, 2026

Source label: grok/grok-3-mini-fast-high-reasoning

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 63.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

83.1%82.3% - 84%

#35 · GPT-4.1

VALS-AI · Jun 17, 2026

Source label: openai/gpt-4.1-2025-04-14

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 62.2%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

83.1%82.3% - 83.9%

#36 · Claude Opus 4

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-20250514

verified runtimevariant directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 61.1%
Last updated: recent
Eligibility: historical_model
Identity: dated variant (0.80)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

83.1%82.2% - 83.9%

#37 · Claude Sonnet 4

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-sonnet-4-20250514

verified runtimevariant directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 60%
Last updated: recent
Eligibility: historical_model
Identity: dated variant (0.80)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

83%82.2% - 83.7%

#38 · GPT-5.2

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.2-2025-12-11

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 58.9%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

82.8%82% - 83.5%

#39 · Grok 3

VALS-AI · Jun 17, 2026

Source label: grok/grok-3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 57.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

82.6%81.8% - 83.4%

#40 · Claude Sonnet 3.7

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-3-7-sonnet-20250219-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 56.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

82.5%81.5% - 83.6%

#41 · Grok 4.1 Fast

VALS-AI · Jun 17, 2026

Source label: grok/grok-4-1-fast-reasoning

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 55.6%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

82.5%81.6% - 83.3%

#42 · GPT-4o

VALS-AI · Jun 17, 2026

Source label: openai/gpt-4o-2024-11-20

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 54.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

82.2%81.3% - 83.1%

#43 · Claude Sonnet 4.6

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-sonnet-4-6

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 53.3%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

82.1%81.2% - 83.1%

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

VALS-AI · Jun 17, 2026

Source label: nvidia/nemotron-3-ultra-550b-a55b

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 52.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Nvidia.

82.1%81.2% - 83%

#45 · Gemini 2.5 Flash-Lite

VALS-AI · Jun 17, 2026

Source label: google/gemini-2.5-flash-lite-preview-09-2025-thinking

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 51.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

82%81.2% - 82.9%

#46 · Qwen3 Max

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3-max

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 50%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Alibaba.

81.9%81.1% - 82.6%

#47 · GPT-5.4 mini

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5-mini-2025-08-07

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 48.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

81.8%80.9% - 82.7%

#48 · kimi-k2-instruct

VALS-AI · Jun 17, 2026

Source label: together/moonshotai/Kimi-K2-Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 47.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Together AI.

81.5%80.6% - 82.3%

#49 · Claude Haiku 4.5

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-haiku-4-5-20251001-thinking

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 46.7%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

81.2%80.3% - 82.2%

#50 · deepseek-v3

VALS-AI · Jun 17, 2026

Source label: fireworks/deepseek-v3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 45.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Fireworks AI.

80.8%79.9% - 81.6%

#51 · Grok 4 Fast

VALS-AI · Jun 17, 2026

Source label: grok/grok-4-fast-reasoning

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 44.4%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

80.6%79.7% - 81.5%

#52 · GPT-4 Turbo

VALS-AI · Jun 17, 2026

Source label: openai/gpt-4-turbo

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 43.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

80.5%79.5% - 81.4%

#53 · o1

VALS-AI · Jun 17, 2026

Source label: openai/o1-2024-12-17

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 42.2%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

80.4%79.3% - 81.5%

#54 · deepseek-v4-pro

VALS-AI · Jun 17, 2026

Source label: deepseek/deepseek-v4-pro

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 41.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: DeepSeek.

80.3%79.4% - 81.2%

#55 · Kimi K2 Thinking

VALS-AI · Jun 17, 2026

Source label: kimi/kimi-k2-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 40%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Moonshot AI.

80.2%79.3% - 81.1%

#56 · Qwen3 235B A22B

VALS-AI · Jun 17, 2026

Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 38.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Fireworks AI.

80.2%79.3% - 81%

#57 · minimax-m2.5

VALS-AI · Jun 17, 2026

Source label: minimax/MiniMax-M2.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 37.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: MiniMax.

80%79% - 80.9%

#58 · command-a-03-2025

VALS-AI · Jun 17, 2026

Source label: cohere/command-a-03-2025

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 36.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Cohere.

79.7%78.9% - 80.5%

#59 · glm-4.6

VALS-AI · Jun 17, 2026

Source label: zai/glm-4.6

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 35.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

79.6%78.6% - 80.6%

#60 · o4 mini

VALS-AI · Jun 17, 2026

Source label: openai/o4-mini-2025-04-16

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 34.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

79.2%78.2% - 80.2%

#61 · Mistral Large (Feb '24)

VALS-AI · Jun 17, 2026

Source label: mistralai/mistral-large-2512

verified runtimevariant directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 33.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: dated variant (0.80)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Mistral AI.

79.1%78.2% - 80%

#62 · mimo-v2.5

VALS-AI · Jun 17, 2026

Source label: xiaomi/mimo-v2.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 32.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Xiaomi.

78.9%77.9% - 79.9%

#63 · Gemini 2.0 Flash

VALS-AI · Jun 17, 2026

Source label: google/gemini-2.0-flash-001

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 31.1%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

78.4%76.8% - 79.9%

#64 · GPT-4.1 mini

VALS-AI · Jun 17, 2026

Source label: openai/gpt-4.1-mini-2025-04-14

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 30%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

78%77.2% - 78.9%

#65 · GPT-5.4 nano

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.4-nano-2026-03-17

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 28.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

77.9%77.1% - 78.8%

#66 · Grok 4.20

VALS-AI · Jun 17, 2026

Source label: grok/grok-4.20-0309-reasoning

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 27.8%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: xAI.

77.7%76.8% - 78.7%

#67 · deepseek-v3-0324

VALS-AI · Jun 17, 2026

Source label: fireworks/deepseek-v3-0324

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 26.7%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Fireworks AI.

77.7%76.9% - 78.5%

#68 · mimo-v2.5-pro

VALS-AI · Jun 17, 2026

Source label: xiaomi/mimo-v2.5-pro

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 25.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Xiaomi.

77.1%76.1% - 78.2%

#69 · GPT-OSS 120B

VALS-AI · Jun 17, 2026

Source label: fireworks/gpt-oss-120b

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 24.4%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Fireworks AI.

75.9%74.9% - 77%

#70 · glm-4.5

VALS-AI · Jun 17, 2026

Source label: zai/glm-4.5

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 23.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

75.6%74.6% - 76.6%

#71 · MiniMax-M2.1

VALS-AI · Jun 17, 2026

Source label: minimax/MiniMax-M2.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 22.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: MiniMax.

75.4%74.4% - 76.5%

#72 · jamba-1.5-large

VALS-AI · Jun 17, 2026

Source label: ai21labs/jamba-1.5-large

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 21.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: AI21 Labs.

74.2%72.9% - 75.4%

#73 · llama-4-scout-17b-16e-instruct

VALS-AI · Jun 17, 2026

Source label: together/meta-llama/Llama-4-Scout-17B-16E-Instruct

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 20%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Together AI.

72%71.1% - 73%

#74 · o3 mini

VALS-AI · Jun 17, 2026

Source label: openai/o3-mini-2025-01-31

verified runtimeexact directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 18.9%
Last updated: recent
Eligibility: historical_model
Identity: exact (1.00)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

71.5%70.1% - 73%

#75 · GPT-OSS 20B

VALS-AI · Jun 17, 2026

Source label: fireworks/gpt-oss-20b

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 17.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Fireworks AI.

70.8%69.7% - 72%

#76 · Claude Haiku 3.5

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-3-5-haiku-20241022

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 16.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Anthropic.

70.3%69.1% - 71.6%

#77 · Gemini 1.0 Pro

VALS-AI · Jun 17, 2026

Source label: google/gemini-1.0-pro-002

verified runtimevariant directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 15.6%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: dated variant (0.80)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

70.2%68.7% - 71.7%

#78 · Mistral Small (Sep '24)

VALS-AI · Jun 17, 2026

Source label: mistralai/mistral-small-2503

verified runtimevariant directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 14.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: dated variant (0.80)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Mistral AI.

69.2%68.2% - 70.2%

#79 · Gemini 1.5 Pro

VALS-AI · Jun 17, 2026

Source label: google/gemini-1.5-pro-002

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 13.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

69.1%67.6% - 70.6%

#80 · command-r-plus

VALS-AI · Jun 17, 2026

Source label: cohere/command-r-plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 12.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Cohere.

68.3%67% - 69.6%

#81 · gemma-2-9b-it

VALS-AI · Jun 17, 2026

Source label: together/google/gemma-2-9b-it

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 11.1%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Together AI.

68.1%66.7% - 69.6%

#82 · gemma-2-27b-it

VALS-AI · Jun 17, 2026

Source label: together/google/gemma-2-27b-it

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 10%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Together AI.

68%66.7% - 69.3%

#83 · DeepSeek Reasoner

VALS-AI · Jun 17, 2026

Source label: fireworks/deepseek-r1

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 8.9%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Fireworks AI.

67.3%66.3% - 68.4%

#84 · jamba-1.5-mini

VALS-AI · Jun 17, 2026

Source label: ai21labs/jamba-1.5-mini

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 7.8%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: AI21 Labs.

66.6%65.3% - 67.9%

#85 · GPT-3.5 Turbo

VALS-AI · Jun 17, 2026

Source label: openai/gpt-3.5-turbo

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 6.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

64.4%63.1% - 65.6%

#86 · Gemini 1.5 Flash

VALS-AI · Jun 17, 2026

Source label: google/gemini-1.5-flash-002

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 5.6%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Google.

63.2%61.8% - 64.7%

#87 · mistral-medium-2505

VALS-AI · Jun 17, 2026

Source label: mistralai/mistral-medium-2505

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 4.4%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Mistral AI.

62%61.1% - 62.9%

#88 · GPT-4.1 nano

VALS-AI · Jun 17, 2026

Source label: openai/gpt-4.1-nano-2025-04-14

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 3.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: OpenAI.

61.1%59.9% - 62.2%

#89 · Magistral Medium 1.2

VALS-AI · Jun 17, 2026

Source label: mistralai/magistral-medium-2509

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 2.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Mistral AI.

54.8%53.7% - 55.9%

#90 · Magistral Small 1.2

VALS-AI · Jun 17, 2026

Source label: mistralai/magistral-small-2509

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 1.1%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Mistral AI.

40%39.1% - 40.9%

#91 · command-r

VALS-AI · Jun 17, 2026

Source label: cohere/command-r

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/legal_bench
Percentile: 0%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Cohere.

33%31.6% - 34.4%

LegalBench

Test details

What it measures vs what it misses

✓ Measures

✗ Misses

Leaderboard · this benchmark version

Loading benchmark evidence.

LegalBench

Test details

What it measures vs what it misses

✓ Measures

✗ Misses

Leaderboard · this benchmark version