Harvey's Legal Agent Benchmark leaderboard
As of 6 Nov 2026, the top score on Harvey's Legal Agent Benchmark is 27.1% by Gemini 4 Argon (Google), ahead of Muse Spark Open at 25.0%, across 34 model configurations with a published result (Vals AI, result dated 5 Nov 2026). Legal tasks that require working across a set of documents, as an associate would, from drafting to review. Harvey's Legal Agent Benchmark is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 6 Nov 2026)
- Gemini 4 Argon holds the top Harvey's Legal Agent Benchmark v1.1 score at 27.1% (Vals AI, 5 Nov 2026).
- Muse Spark Open is second on Harvey's Legal Agent Benchmark v1.1 at 25.0% (Vals AI, 5 Nov 2026).
- Harvey's Legal Agent Benchmark v1.1: Agentic legal work over files, written with Harvey. Run by Vals AI (Vals AI).
- 34 model configurations have a published Harvey's Legal Agent Benchmark v1.1 result on BenchLeader (BenchLeader, 6 Nov 2026).
- Harvey's Legal Agent Benchmark v1.1 does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
Agentic legal work over files, written with Harvey. Run by Vals AI.
- Published by
- Vals AI
- Category
- Agents & tools
- Index weight
- Reference only
- Models
- 34
- Data as of
- 6 Nov 2026
Vals AI (vals.ai).
What the test looks like
Legal tasks that require working across a set of documents, as an associate would, from drafting to review.
How it is scored
Rubric score, run by Vals AI with Harvey.
What to keep in mind
Co-designed with a legal-AI vendor, which Vals discloses.
- 1Gemini 4 Argon (high)27.1%
- 2Muse Spark Open (xhigh)25.0%
- 3Muse Wonder (max)23.8%
- 4Muse Wonder Full (max)23.3%
- 5Inkling 2 Small21.7%
- 6Grok 4.7 (xhigh)18.3%
- 7Gemini 4 Carbon (high)17.5%
- 8Mistral Large 4 (high)17.5%
- 9Mistral Large 4 OSS (high)15.8%
- 10Qwen3.8 Max Internal15.4%
- 11GLM 5.3 (max)14.6%
- 12DeepSeek V4.1 Flash 1017 (max)14.2%
- 13MiMo-V2.6-Pro13.8%
- 14MiMo-V2.6-Flash13.3%
- 15Kimi K3 (max)12.9%
34 of 34
| # | |||||
|---|---|---|---|---|---|
| 1 |  Gemini 4 Argonhigh | 27.1% | 66.0 | $10.55 | 2026-11-05 |
| 2 |  Muse Spark Openxhighopen ↗ | 25.0% | 54.0 | $13.63 | 2026-11-05 |
| 3 | Muse Wondermax | 23.8% | 62.3 | $7.08 | 2026-11-05 |
| 4 | Muse Wonder Fullmax | 23.3% | 62.9 | $5.41 | 2026-11-05 |
| 5 | NewInkling 2 Small | 21.7% | 57.4 | $1.23 | 2026-11-05 |
| 6 |  Grok 4.7xhigh | 18.3% | 58.3 | $10.75 | 2026-11-05 |
| 7 | Gemini 4 Carbonhigh | 17.5% | 69.2 | $6.52 | 2026-11-05 |
| 8 |  Mistral Large 4highopen ↗ | 17.5% | 49.5 | $2.55 | 2026-11-05 |
| 9 | Mistral Large 4 OSShigh | 15.8% | – | $0.733 | 2026-11-05 |
| 10 |  Qwen3.8 Max Internal | 15.4% | – | $3.36 | 2026-11-05 |
| 11 |  GLM 5.3maxopen ↗ | 14.6% | 59.0 | $3.99 | 2026-11-05 |
| 12 |  DeepSeek V4.1 Flash 1017maxopen ↗ | 14.2% | 59.2 | $0.613 | 2026-11-05 |
| 13 |  MiMo-V2.6-Proopen ↗ | 13.8% | 57.3 | $0.217 | 2026-11-05 |
| 14 | MiMo-V2.6-Flashopen ↗ | 13.3% | 55.0 | $0.086 | 2026-11-05 |
| 15 |  Kimi K3maxopen ↗ | 12.9% | 62.1 | $4.14 | 2026-11-05 |
| 16 |  Step 5 Previewopen | 12.9% | 54.9 | $1.13 | 2026-11-05 |
| 17 |  Hy4open ↗ | 12.5% | 55.7 | $1.04 | 2026-11-05 |
| 18 | Gemini 3.8 Flashhigh | 12.1% | 62.0 | $3.66 | 2026-11-05 |
| 19 | Kimi K3.1 Promaxopen ↗ | 11.7% | – | $9.00 | 2026-11-05 |
| 20 | DeepSeek V4 Promaxopen ↗ | 10.8% | 57.0 | $0.802 | 2026-11-05 |
| 21 | GLM 5.3 Flashmaxopen ↗ | 10.4% | 54.2 | $0.323 | 2026-11-05 |
| 22 |  Claude Sonnet 5.5 | 10.4% | – | $16.61 | 2026-11-05 |
| 23 | Claude Opus 5.5 | 9.2% | 63.4 | $21.38 | 2026-11-05 |
| 24 |  MiniMax M3.1 Flashopen ↗ | 7.5% | 53.0 | $0.473 | 2026-11-05 |
| 25 |  GPT-6 Astramax | 7.1% | 67.8 | $25.46 | 2026-11-05 |
| 26 | GPT-6.1 Solmax | 6.3% | 67.3 | $3.54 | 2026-11-05 |
| 27 |  Mercury 3high | 5.0% | – | $0.188 | 2026-11-05 |
| 28 |  Beam 1.0open | 4.6% | 45.8 | – | 2026-11-05 |
| 29 | GPT-6 Lunamax | 4.2% | 56.7 | $0.298 | 2026-11-05 |
| 30 | Claude Haiku 5.5 | 4.2% | – | $4.26 | 2026-11-05 |
| 31 | Inklingopen ↗ | 3.8% | 47.5 | $1.25 | 2026-11-05 |
| 32 |  Nemotron 3.5 Super VL 120B A12Bhighopen | 3.8% | – | $0.373 | 2026-11-05 |
| 33 | Inkling Smallopen ↗ | 2.9% | 52.1 | $0.179 | 2026-11-05 |
| 34 | GPT-6.1 Terramax | 1.7% | 65.3 | $1.48 | 2026-11-05 |
More agents & tools benchmarks · All LLM benchmarks
Related benchmarks
Top head-to-heads on Harvey's Legal Agent Benchmark v1.1
Cite as: BenchLeader, “Harvey's Legal Agent Benchmark v1.1 leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/vals_hlab.
Harvey's Legal Agent Benchmark v1.1: questions
- What does Harvey's Legal Agent Benchmark v1.1 measure?
- Legal tasks that require working across a set of documents, as an associate would, from drafting to review. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on Harvey's Legal Agent Benchmark?
- As of 6 Nov 2026, the highest score on Harvey's Legal Agent Benchmark v1.1 is 27.1% by Gemini 4 Argon (result dated 5 Nov 2026), ahead of Muse Spark Open at 25.0%. Scores come from Vals AI and are re-read daily.
- Which AI model leads Harvey's Legal Agent Benchmark v1.1?
- Gemini 4 Argon leads Harvey's Legal Agent Benchmark v1.1 with 27.1% as of 6 Nov 2026 (result dated 5 Nov 2026), ahead of Muse Spark Open at 25.0%.
- How many models have Harvey's Legal Agent Benchmark v1.1 results?
- 34 model configurations have a Harvey's Legal Agent Benchmark v1.1 result on BenchLeader, all taken from Vals AI.
- Who runs Harvey's Legal Agent Benchmark v1.1 and how often is it updated?
- Harvey's Legal Agent Benchmark v1.1 is published by Vals AI. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does Harvey's Legal Agent Benchmark v1.1 count toward the BenchLeader Index?
- No. Harvey's Legal Agent Benchmark v1.1 is shown for reference but left out of the composite index.