Benchmarks · agentic
Legal Research Bench
Evaluating agents on legal research tasks across diverse areas of US law
65 models · 66 results · updated 2026-11-05 · vals.ai ↗
Leaderboard
Gemini 4 Carbon high
Heavy line: best score at each price. Thin lines join one model's effort levels. Cost uses each model's median subscription plan, or API price with no plan data. 50 more models under All models in settings.
More from this benchmark
1 published measuresTime per task91 results · lower is better
45 more models under All models in settings.
All results
66 rows| # | Model | Effort | Legal Research Bench | Subscription $ / task | API $ / task | Time |
|---|---|---|---|---|---|---|
| 1 | Gemini 4 Carbon | high | 55.8% | $3.12 | $8.74 | 52 min |
| 2 | Fable 5.1 | max | 55.3% | $3.59 | $23 | 57 min |
| 3 | Opus 5 | max | 55.3% | $0.12 | $6.76 | 31 min |
| 4 | Muse Spark 1.3 | max | 55.3% | $0.29 | $0.59 | 5.3 min |
| 5 | Gemini 4 Argon | high | 54.8% | $4.60 | $6.45 | 31 min |
| 6 | Opus 5.5 | max | 50.5% | $0.38 | $25 | 1.9 h |
| 7 | Fable 5 | max | 49.5% | $1.10 | $9.79 | 22 min |
| 8 | GLM 5.3 | max | 49.0% | $0.23 | $2.24 | 52 min |
| 9 | Sonnet 5.5 | max | 48.1% | $0.22 | $15 | 1.6 h |
| 10 | GPT-5.6 Sol | max | 48.1% | $0.60 | $22 | 1.3 h |
| 11 | Grok 4.6 | high | 48.1% | $0.16 | $1.53 | 25 min |
| 12 | Qwen 3.8 Max | max | 47.6% | $0.29 | $2.49 | 1.0 h |
| 13 | Grok 4.7 | xhigh | 47.1% | $0.46 | $4.92 | 21 min |
| 14 | MiMo V2.6 Pro | max | 47.1% | $0.15 | $0.18 | 30 min |
| 15 | Inkling 2 Small | max | 46.6% | $0.29 | $0.29 | 12 min |
| 16 | Kimi K3 | max | 46.2% | $0.32 | $3.47 | 13 min |
| 17 | DeepSeek V4.1 Flash 1017 | max | 45.2% | free | $0.70 | 35 min |
| 18 | GLM 5.3 Flash | max | 45.2% | free | $0.24 | 51 min |
| 19 | Hy4 preview | high | 45.2% | $0.64 | $0.85 | 1.1 h |
| 20 | Haiku 5.5 | max | 43.8% | $0.035 | $1.65 | 1.4 h |
| 21 | Opus 4.8 | max | 43.8% | $0.049 | $2.82 | 12 min |
| 22 | Muse Spark 1.2 | xhigh | 43.8% | $0.25 | $0.51 | 5.1 min |
| 23 | Sonnet 5 | max | 41.8% | $0.047 | $2.72 | 26 min |
| 24 | DeepSeek V4.1 Flash | max | 41.3% | $0.018 | $0.25 | 12 min |
| 25 | GPT-5.6 Terra | max | 41.3% | $0.22 | $7.91 | 1.2 h |
| 26 | Hy4 | high | 41.3% | $0.64 | $2.22 | 1.5 h |
| 27 | DeepSeek V4 Pro 0813 | max | 40.9% | $1.09 | $1.09 | 42 min |
| 28 | Muse Spark 1.3 | xhigh | 40.9% | $0.33 | $0.65 | 8.8 min |
| 29 | GPT-5.5 | xhigh | 40.4% | $0.20 | $7.40 | 38 min |
| 30 | GPT-6 Astra | max | 39.4% | $0.45 | $11 | 25 min |