Legal Research Bench

Agentic legal research across areas of US law. Run by Vals AI.

CategoryKnowledge
MetricAccuracy (%)
SourceVals AI (evaluator)
LicenseNone stated
Benchmark pagewww.vals.ai
StatusScored
Weight0.8 within Knowledge
FitDiscrimination 0.60, noise 0.39

Results

#ModelProviderResultDate
1 Gemini 4 Carbon High Google 55.8% 2026-11-09
2 Claude Opus 5 Max Anthropic 55.3% 2026-11-09
3 Claude Fable 5.1 Max Anthropic 55.3% 2026-11-09
4 Muse Spark 1.3 Max Meta 55.3% 2026-11-09
5 Gemini 4 Argon High Google 54.8% 2026-11-09
6 Muse Wonder Full Max Meta 53.4% 2026-11-09
7 Claude Opus 5.5 Max Anthropic 50.5% 2026-11-09
8 Muse Wonder Max Meta 50.5% 2026-11-09
9 Claude Fable 5 Max Anthropic 49.5% 2026-11-09
10 GLM-5.3 Max Zhipu AI 49.0% 2026-11-09
11 Muse Ember Max Meta 49.0% 2026-11-09
12 Claude Sonnet 5.5 Max Anthropic 48.1% 2026-11-09
13 Grok 4.6 High xAI 48.1% 2026-11-09
14 GPT-5.6 Sol Max OpenAI 48.1% 2026-11-09
15 Qwen3.8 Max Alibaba 47.6% 2026-11-09
16 MiMo-V2.6-Pro Xiaomi 47.1% 2026-11-09
17 Grok 4.7 Extra High xAI 47.1% 2026-11-09
18 Inkling 2 Small Thinking Machines 46.6% 2026-11-09
19 Kimi K3 Max Moonshot AI 46.2% 2026-11-09
20 Claude Fable 5.2 High Anthropic 46.2% 2026-11-09
21 Hy4 preview Tencent 45.2% 2026-11-09
22 DeepSeek V4.1 Flash 1017 Max DeepSeek 45.2% 2026-11-09
23 GLM-5.3-Flash Max Zhipu AI 45.2% 2026-11-09
24 Ember-1 Fireworks 44.7% 2026-11-09
25 Claude Haiku 5.5 Max Anthropic 43.8% 2026-11-09
26 Claude Opus 4.8 Max Anthropic 43.8% 2026-11-09
27 Muse Spark 1.2 Extra High Meta 43.8% 2026-11-09
28 Claude Sonnet 5 Max Anthropic 41.8% 2026-11-09
29 Hy4 Tencent 41.3% 2026-11-09
30 DeepSeek V4.1 Flash Max DeepSeek 41.3% 2026-11-09
31 GPT-5.6 Terra Max OpenAI 41.3% 2026-11-09
32 DeepSeek V4 Pro 0813 Max DeepSeek 40.9% 2026-11-09
33 Muse Spark 1.3 Extra High Meta 40.9% 2026-11-09
34 Step 5 Preview StepFun 40.4% 2026-11-09
35 GPT-5.5 Extra High OpenAI 40.4% 2026-11-09
36 GPT-6 Astra Max OpenAI 39.4% 2026-11-09
37 Gemini 3.8 Flash High Google 38.9% 2026-11-09
38 Muse Spark Open Extra High Meta 38.5% 2026-11-09
39 GPT-6.1 Sol Max OpenAI 38.5% 2026-11-09
40 Claude Sonnet 4.6 Max Anthropic 38.5% 2026-11-09
41 Claude Opus 4.7 Max Anthropic 38.5% 2026-11-09
42 MiMo-V2.6-Flash Xiaomi 38.0% 2026-11-09
43 Grok 4.5 High xAI 38.0% 2026-11-09
44 Muse Spark 1.1 Extra High Meta 38.0% 2026-11-09
45 Kimi K3.1 Pro Max Moonshot AI 37.5% 2026-11-09
46 GPT-5.6 Luna Max OpenAI 36.5% 2026-11-09
47 Qwen3.8 27B Extra High Alibaba 36.1% 2026-11-09
48 Gemini 3.7 Flash High Google 34.6% 2026-11-09
49 MiniMax-M3.1-Flash Max MiniMax 34.1% 2026-11-09
50 Mistral Large 4 High Mistral 31.7% 2026-11-09
51 GLM-5.2 Max Zhipu AI 31.3% 2026-11-09
52 Gemini 3.5 Flash High Google 30.8% 2026-11-09
53 GPT-6 Luna Max OpenAI 30.3% 2026-11-09
54 DeepSeek V4 Flash 0731 High DeepSeek 30.3% 2026-11-09
55 MiniMax-M3 MiniMax 29.8% 2026-11-09
56 Beam 1.0 Reflection 28.8% 2026-11-09
57 GPT-6 Sol Max OpenAI 28.8% 2026-11-09
58 Mercury 3 High Inception 28.8% 2026-11-09
59 Inkling Thinking Machines 28.4% 2026-11-09
60 GLM-5.1 Zhipu AI 27.9% 2026-11-09
61 GPT-6.1 Terra Max OpenAI 26.0% 2026-11-09
62 Inkling Small Thinking Machines 25.5% 2026-11-09
63 Qwen3.7 Max Alibaba 25.5% 2026-11-09
64 Gemini 3.6 Flash High Google 25.0% 2026-11-09
65 DeepSeek V4 Pro Max DeepSeek 23.1% 2026-11-09
66 Gemini 3.1 Pro Preview High Google 20.7% 2026-11-09
67 Gemini 3 Flash Preview High Google 18.3% 2026-11-09
68 Qwen3.7 Plus Alibaba 16.3% 2026-11-09
69 Kimi K2.6 Moonshot AI 15.9% 2026-11-09
70 MiMo-V2.5-Pro Xiaomi 15.9% 2026-11-09
71 Kimi K2.5 Thinking Moonshot AI 15.9% 2026-11-09
72 Nemotron 3 Ultra 550B A55B NVIDIA 15.4% 2026-11-09
73 Grok 4.3 High xAI 15.4% 2026-11-09
74 Qwen3.6 Plus Alibaba 14.9% 2026-11-09
75 Grok 4.20 (Reasoning) xAI 13.9% 2026-11-09
76 Gemini 3.5 Flash Lite High Google 13.9% 2026-11-09
77 GPT-5.4 mini Extra High OpenAI 12.5% 2026-11-09
78 MiniMax-M2.7 MiniMax 10.6% 2026-11-09
79 Claude Haiku 4.5 Thinking Anthropic 10.6% 2026-11-09
80 Nemotron 3.5 Super VL 120B A12B High NVIDIA 10.1% 2026-11-09
81 MiMo-V2.5 Xiaomi 9.1% 2026-11-09
82 Mistral Medium 3.5 High Mistral 9.1% 2026-11-09
83 GPT-5.4 nano High OpenAI 6.3% 2026-11-09
84 Mercury 2.5 High Inception 4.3% 2026-11-09
85 Gemini 3.1 Flash Lite Preview High Google 3.4% 2026-11-09
86 Laguna M.1 Poolside 2.4% 2026-11-09
87 Laguna XS.2 Poolside 1.0% 2026-11-09