Harvey LAB v1.1 (HLAB-Vals-1.1)
Agents produce legal work product from matter files with file tools and docx, pptx and xlsx skills. A task passes only if every expert rubric criterion passes.
Harvey fixed the 120-task private holdout, removing 17 unsolvable criteria. The harness was resynced with a finish tool and a reasoning-first judge. Vals says scores are not comparable with the v1 board.
- Made by
- Harvey (board run by Vals AI)
- Released
- October 2026
- Kind
- Law
- Level
- Expert
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- Public, plus a private test set
- Measures
- Capability
- Subject
- Law
- Language
- English
- Interaction
- Agentic
- Environment
- Terminal
- Items from
- Real work and use, Another benchmark
- Refresh
- New versions replace it
- Run by
- A host
- Based on
- Harvey LAB (Legal Agent Benchmark)
- Occupations
- Legal
- In views
- Agents, Law and finance
- Strongest model tested
- GPT-6.2 Astra #1 on the capability index
- Last in a launch post
- November 2026
- Last model added
- November 2026
- Scores refreshed
- November 19th, 2026
Gemini 4 Argon leads with 27.1% Gemini 4 Argon leads with 27.1%
Best score per entry from the benchmark's own leaderboard, read on November 19th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
Gemini 4 Argon leads with 27.1%, claims included
Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model. Striped bars are claims, and dotted bars are reruns by Epoch AI or another board.
- 27.1
Gemini 4 Argon
- 25.0
Muse Spark Open
- 25.0
Muse Ember
- 23.8
Muse Wonder
- 23.3
Muse Wonder Full
- 21.7
Inkling 2 Small
- 21.3
Qwen4-Max
- 19.6
Gemini 4 Neon
- 18.3 Grok 4.7
- 17.5
Mistral Large 4 Preview
- 17.5
Gemini 4 Carbon
- 16.3 GPT-6.2 Astra
- 15.8
Mistral Large 4
- 15.4
Qwen3.8 Max (0902)
- 14.6 GLM-5.3
- 14.2
DeepSeek V4.1 Flash 1017
- 13.8 MiMo-V2.6 Pro
- 13.3 MiMo-V2.6 Flash
- 12.9 Kimi K3
- 12.9
Step 5 Preview
- 12.5
Tencent Hy4
- 12.1
Gemini 3.8 Flash
- 12.1 GPT-6.2 Sol
- 11.7 Kimi K3.1 Pro
- 11.3
Gemini 4 Hydrogen
- 10.8
DeepSeek V4 Pro 0813
- 10.4 GLM-5.3-Flash
- 10.4 Claude Sonnet 5.5
- 9.17 Claude Opus 5.5
- 8.75
Tencent Hy4 preview
- 7.50
MiniMax-M3.1-Flash
- 7.08 GPT-6 Astra
- 6.25 GPT-6.1 Sol
- 5.42 Hy4 Flash
- 5.00 Mercury 3
- 5.00 Claude Fable 5.2
- 4.58 Beam 1.0
- 4.17 GPT-6 Luna
- 4.17 Claude Haiku 5.5
- 3.75
Inkling
- 3.75
Nemotron 3.5 Super VL
- 2.92
Inkling-Small
- 1.67 GPT-6.1 Terra
- 1
Gemini 4 Argon 27.1%
- 2
Muse Spark Open 25%
- 3
Muse Ember 25%
- 4
Muse Wonder 23.8%
- 5
Muse Wonder Full 23.3%
- 6
Inkling 2 Small 21.7%
- 7
Qwen4-Max 21.3%
- 8
Gemini 4 Neon 19.6%
- 9 Grok 4.7 18.3%
- 10
Mistral Large 4 Preview 17.5%
- 11
Gemini 4 Carbon 17.5%
- 12 GPT-6.2 Astra 16.3%
- 13
Mistral Large 4 15.8%
- 14
Qwen3.8 Max (0902) 15.4%
- 15 GLM-5.3 14.6%
- 16
DeepSeek V4.1 Flash 1017 14.2%
- 17 MiMo-V2.6 Pro 13.8%
- 18 MiMo-V2.6 Flash 13.3%
- 19 Kimi K3 12.9%
- 20
Step 5 Preview 12.9%
- 21
Tencent Hy4 12.5%
- 22
Gemini 3.8 Flash 12.1%
- 23 GPT-6.2 Sol 12.1%
- 24 Kimi K3.1 Pro 11.7%
- 25
Gemini 4 Hydrogen 11.3%
- 26
DeepSeek V4 Pro 0813 10.8%
- 27 GLM-5.3-Flash 10.4%
- 28 Claude Sonnet 5.5 10.4%
- 29 Claude Opus 5.5 9.17%
- 30
Tencent Hy4 preview 8.75%
- 31
MiniMax-M3.1-Flash 7.50%
- 32 GPT-6 Astra 7.08%
- 33 GPT-6.1 Sol 6.25%
- 34 Hy4 Flash 5.42%
- 35 Mercury 3 5.00%
- 36 Claude Fable 5.2 5.00%
- 37 Beam 1.0 4.58%
- 38 GPT-6 Luna 4.17%
- 39 Claude Haiku 5.5 4.17%
- 40
Inkling 3.75%
- 41
Nemotron 3.5 Super VL 3.75%
- 42
Inkling-Small 2.92%
- 43 GPT-6.1 Terra 1.67%
Source: Official Harvey LAB v1.1 (HLAB-Vals-1.1) leaderboard
- 27.1
Gemini 4 Argon
- 25.0
Muse Spark Open
- 25.0
Muse Ember
- 23.8
Muse Wonder
- 23.3
Muse Wonder Full
- 21.7
Inkling 2 Small
- 21.3
Qwen4-Max
- 19.6
Gemini 4 Neon
- 18.3 Grok 4.7
- 17.5
Mistral Large 4 Preview
- 17.5
Gemini 4 Carbon
- 16.3 GPT-6.2 Astra
- 15.8
Mistral Large 4
- 15.4
Qwen3.8 Max (0902)
- 14.6 GLM-5.3
- 14.2
DeepSeek V4.1 Flash 1017
- 13.8 MiMo-V2.6 Pro
- 13.3 MiMo-V2.6 Flash
- 12.9 Kimi K3
- 12.9
Step 5 Preview
- 12.5
Tencent Hy4
- 12.1
Gemini 3.8 Flash
- 12.1 GPT-6.2 Sol
- 11.7 Kimi K3.1 Pro
- 11.3
Gemini 4 Hydrogen
- 10.8
DeepSeek V4 Pro 0813
- 10.4 GLM-5.3-Flash
- 10.4 Claude Sonnet 5.5
- 9.17 Claude Opus 5.5
- 8.75
Tencent Hy4 preview
- 7.50
MiniMax-M3.1-Flash
- 7.08 GPT-6 Astra
- 6.25 GPT-6.1 Sol
- 5.00 Mercury 3
- 5.00 Claude Fable 5.2
- 4.58 Beam 1.0
- 4.17 GPT-6 Luna
- 4.17 Claude Haiku 5.5
- 3.75
Inkling
- 3.75
Nemotron 3.5 Super VL
- 2.92
Inkling-Small
- 1.67 GPT-6.1 Terra
- 1
Gemini 4 Argon 27.1%
- 2
Muse Spark Open 25%
- 3
Muse Ember 25%
- 4
Muse Wonder 23.8%
- 5
Muse Wonder Full 23.3%
- 6
Inkling 2 Small 21.7%
- 7
Qwen4-Max 21.3%
- 8
Gemini 4 Neon 19.6%
- 9 Grok 4.7 18.3%
- 10
Mistral Large 4 Preview 17.5%
- 11
Gemini 4 Carbon 17.5%
- 12 GPT-6.2 Astra 16.3%
- 13
Mistral Large 4 15.8%
- 14
Qwen3.8 Max (0902) 15.4%
- 15 GLM-5.3 14.6%
- 16
DeepSeek V4.1 Flash 1017 14.2%
- 17 MiMo-V2.6 Pro 13.8%
- 18 MiMo-V2.6 Flash 13.3%
- 19 Kimi K3 12.9%
- 20
Step 5 Preview 12.9%
- 21
Tencent Hy4 12.5%
- 22
Gemini 3.8 Flash 12.1%
- 23 GPT-6.2 Sol 12.1%
- 24 Kimi K3.1 Pro 11.7%
- 25
Gemini 4 Hydrogen 11.3%
- 26
DeepSeek V4 Pro 0813 10.8%
- 27 GLM-5.3-Flash 10.4%
- 28 Claude Sonnet 5.5 10.4%
- 29 Claude Opus 5.5 9.17%
- 30
Tencent Hy4 preview 8.75%
- 31
MiniMax-M3.1-Flash 7.50%
- 32 GPT-6 Astra 7.08%
- 33 GPT-6.1 Sol 6.25%
- 34 Mercury 3 5.00%
- 35 Claude Fable 5.2 5.00%
- 36 Beam 1.0 4.58%
- 37 GPT-6 Luna 4.17%
- 38 Claude Haiku 5.5 4.17%
- 39
Inkling 3.75%
- 40
Nemotron 3.5 Super VL 3.75%
- 41
Inkling-Small 2.92%
- 42 GPT-6.1 Terra 1.67%
Source: Official Harvey LAB v1.1 (HLAB-Vals-1.1) leaderboard
- 27.1
Gemini 4 Argon
- 25.0
Muse Spark Open
- 25.0
Muse Ember
- 23.8
Muse Wonder
- 23.3
Muse Wonder Full
- 21.7
Inkling 2 Small
- 21.3
Qwen4-Max
- 19.6
Gemini 4 Neon
- 18.3 Grok 4.7
- 17.5
Mistral Large 4 Preview
- 17.5
Gemini 4 Carbon
- 16.3 GPT-6.2 Astra
- 15.8
Mistral Large 4
- 15.4
Qwen3.8 Max (0902)
- 14.6 GLM-5.3
- 14.2
DeepSeek V4.1 Flash 1017
- 13.8 MiMo-V2.6 Pro
- 13.3 MiMo-V2.6 Flash
- 12.9 Kimi K3
- 12.9
Step 5 Preview
- 12.5
Tencent Hy4
- 12.1
Gemini 3.8 Flash
- 12.1 GPT-6.2 Sol
- 11.7 Kimi K3.1 Pro
- 11.3
Gemini 4 Hydrogen
- 10.8
DeepSeek V4 Pro 0813
- 10.4 GLM-5.3-Flash
- 10.4 Claude Sonnet 5.5
- 9.17 Claude Opus 5.5
- 8.75
Tencent Hy4 preview
- 7.50
MiniMax-M3.1-Flash
- 7.08 GPT-6 Astra
- 6.25 GPT-6.1 Sol
- 5.00 Mercury 3
- 5.00 Claude Fable 5.2
- 4.58 Beam 1.0
- 4.17 GPT-6 Luna
- 4.17 Claude Haiku 5.5
- 3.75
Inkling
- 3.75
Nemotron 3.5 Super VL
- 2.92
Inkling-Small
- 1.67 GPT-6.1 Terra
- 1
Gemini 4 Argon 27.1%
- 2
Muse Spark Open 25%
- 3
Muse Ember 25%
- 4
Muse Wonder 23.8%
- 5
Muse Wonder Full 23.3%
- 6
Inkling 2 Small 21.7%
- 7
Qwen4-Max 21.3%
- 8
Gemini 4 Neon 19.6%
- 9 Grok 4.7 18.3%
- 10
Mistral Large 4 Preview 17.5%
- 11
Gemini 4 Carbon 17.5%
- 12 GPT-6.2 Astra 16.3%
- 13
Mistral Large 4 15.8%
- 14
Qwen3.8 Max (0902) 15.4%
- 15 GLM-5.3 14.6%
- 16
DeepSeek V4.1 Flash 1017 14.2%
- 17 MiMo-V2.6 Pro 13.8%
- 18 MiMo-V2.6 Flash 13.3%
- 19 Kimi K3 12.9%
- 20
Step 5 Preview 12.9%
- 21
Tencent Hy4 12.5%
- 22
Gemini 3.8 Flash 12.1%
- 23 GPT-6.2 Sol 12.1%
- 24 Kimi K3.1 Pro 11.7%
- 25
Gemini 4 Hydrogen 11.3%
- 26
DeepSeek V4 Pro 0813 10.8%
- 27 GLM-5.3-Flash 10.4%
- 28 Claude Sonnet 5.5 10.4%
- 29 Claude Opus 5.5 9.17%
- 30
Tencent Hy4 preview 8.75%
- 31
MiniMax-M3.1-Flash 7.50%
- 32 GPT-6 Astra 7.08%
- 33 GPT-6.1 Sol 6.25%
- 34 Mercury 3 5.00%
- 35 Claude Fable 5.2 5.00%
- 36 Beam 1.0 4.58%
- 37 GPT-6 Luna 4.17%
- 38 Claude Haiku 5.5 4.17%
- 39
Inkling 3.75%
- 40
Nemotron 3.5 Super VL 3.75%
- 41
Inkling-Small 2.92%
- 42 GPT-6.1 Terra 1.67%
Sources: Official Harvey LAB v1.1 (HLAB-Vals-1.1) leaderboard and labs’ model cards
| # | Entry | Score ± 95% CI | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 27.1% ±7.3 | Sep 2026 | Official leaderboard | |
| 2 | | 25% ±7.1 | Oct 2026 | Official leaderboard | |
| 3 | | 25% ±7.2 | Nov 2026 | Official leaderboard | |
| 4 | | 23.8% ±7.1 | Oct 2026 | Official leaderboard | |
| 5 | | 23.3% ±7 | Oct 2026 | Official leaderboard | |
| 6 | | 21.7% ±6.5 | Nov 2026 | Official leaderboard | |
| 7 | | 21.3% ±6.5 | Nov 2026 | Official leaderboard | |
| 8 | | 19.6% ±6.5 | Nov 2026 | Official leaderboard | |
| 9 | Grok 4.7 | 18.3% ±6.2 | Sep 2026 | Official leaderboard | |
| 10 | | 17.5% ±6.3 | Oct 2026 | Official leaderboard | |
| 11 | | 17.5% ±5.9 | Oct 2026 | Official leaderboard | |
| 12 | GPT-6.2 Astra | 16.3% ±6 | Nov 2026 | Official leaderboard | |
| 13 | | 15.8% ±6.1 | Oct 2026 | Official leaderboard | |
| 14 | | 15.4% ±6 | Sep 2026 | Official leaderboard | |
| 15 | GLM-5.3 | 14.6% ±5.6 | Aug 2026 | Official leaderboard | |
| 16 | | 14.2% ±5.5 | Oct 2026 | Official leaderboard | |
| 17 | MiMo-V2.6 Pro | 13.8% ±5.6 | Sep 2026 | Official leaderboard | |
| 18 | MiMo-V2.6 Flash | 13.3% ±5.5 | Sep 2026 | Official leaderboard | |
| 19 | Kimi K3 | 12.9% ±5.5 | Jul 2026 | Official leaderboard | |
| 20 | | 12.9% ±5.5 | Sep 2026 | Official leaderboard | |
| 21 | | 12.5% ±5.2 | Nov 2026 | Official leaderboard | |
| 22 | | 12.1% ±5.1 | Sep 2026 | Official leaderboard | |
| 23 | GPT-6.2 Sol | 12.1% ±5.3 | Nov 2026 | Official leaderboard | |
| 24 | Kimi K3.1 Pro | 11.7% ±5.2 | Oct 2026 | Official leaderboard | |
| 25 | | 11.3% ±5.2 | Nov 2026 | Official leaderboard | |
| 26 | | 10.8% ±5.3 | Aug 2026 | Official leaderboard | |
| 27 | GLM-5.3-Flash | 10.4% ±5 | Aug 2026 | Official leaderboard | |
| 28 | Claude Sonnet 5.5 | 10.4% ±5 | Sep 2026 | Official leaderboard | |
| 29 | Claude Opus 5.5 | 9.17% ±4.6 | Sep 2026 | Official leaderboard | |
| 30 | | 8.75% ±4.3 | Aug 2026 | Official leaderboard | |
| 31 | | 7.50% ±4.4 | Oct 2026 | Official leaderboard | |
| 32 | GPT-6 Astra | 7.08% ±3.7 | Sep 2026 | Official leaderboard | |
| 33 | GPT-6.1 Sol | 6.25% ±3.6 | Sep 2026 | Official leaderboard | |
| 34 | Hy4 Flash | 5.42% ±3.6 | Official leaderboard | ||
| 35 | Mercury 3 | 5.00% ±3.2 | Oct 2026 | Official leaderboard | |
| 36 | Claude Fable 5.2 | 5.00% ±3.4 | Nov 2026 | Official leaderboard | |
| 37 | Beam 1.0 | 4.58% ±3.5 | Oct 2026 | Official leaderboard | |
| 38 | GPT-6 Luna | 4.17% ±3.2 | Sep 2026 | Official leaderboard | |
| 39 | Claude Haiku 5.5 | 4.17% ±3 | Oct 2026 | Official leaderboard | |
| 40 | | 3.75% ±2.9 | Jul 2026 | Official leaderboard | |
| 41 | | 3.75% ±2.6 | Oct 2026 | Official leaderboard | |
| 42 | | 2.92% ±2.7 | Jul 2026 | Official leaderboard | |
| 43 | GPT-6.1 Terra | 1.67% ±2 | Oct 2026 | Official leaderboard |
| # | Model | Score ± 95% CI | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 27.1% ±7.3 | Sep 2026 | Official leaderboard | |
| 2 | | 25% ±7.1 | Oct 2026 | Official leaderboard | |
| 3 | | 25% ±7.2 | Nov 2026 | Official leaderboard | |
| 4 | | 23.8% ±7.1 | Oct 2026 | Official leaderboard | |
| 5 | | 23.3% ±7 | Oct 2026 | Official leaderboard | |
| 6 | | 21.7% ±6.5 | Nov 2026 | Official leaderboard | |
| 7 | | 21.3% ±6.5 | Nov 2026 | Official leaderboard | |
| 8 | | 19.6% ±6.5 | Nov 2026 | Official leaderboard | |
| 9 | Grok 4.7 | 18.3% ±6.2 | Sep 2026 | Official leaderboard | |
| 10 | | 17.5% ±6.3 | Oct 2026 | Official leaderboard | |
| 11 | | 17.5% ±5.9 | Oct 2026 | Official leaderboard | |
| 12 | GPT-6.2 Astra | 16.3% ±6 | Nov 2026 | Official leaderboard | |
| 13 | | 15.8% ±6.1 | Oct 2026 | Official leaderboard | |
| 14 | | 15.4% ±6 | Sep 2026 | Official leaderboard | |
| 15 | GLM-5.3 | 14.6% ±5.6 | Aug 2026 | Official leaderboard | |
| 16 | | 14.2% ±5.5 | Oct 2026 | Official leaderboard | |
| 17 | MiMo-V2.6 Pro | 13.8% ±5.6 | Sep 2026 | Official leaderboard | |
| 18 | MiMo-V2.6 Flash | 13.3% ±5.5 | Sep 2026 | Official leaderboard | |
| 19 | Kimi K3 | 12.9% ±5.5 | Jul 2026 | Official leaderboard | |
| 20 | | 12.9% ±5.5 | Sep 2026 | Official leaderboard | |
| 21 | | 12.5% ±5.2 | Nov 2026 | Official leaderboard | |
| 22 | | 12.1% ±5.1 | Sep 2026 | Official leaderboard | |
| 23 | GPT-6.2 Sol | 12.1% ±5.3 | Nov 2026 | Official leaderboard | |
| 24 | Kimi K3.1 Pro | 11.7% ±5.2 | Oct 2026 | Official leaderboard | |
| 25 | | 11.3% ±5.2 | Nov 2026 | Official leaderboard | |
| 26 | | 10.8% ±5.3 | Aug 2026 | Official leaderboard | |
| 27 | GLM-5.3-Flash | 10.4% ±5 | Aug 2026 | Official leaderboard | |
| 28 | Claude Sonnet 5.5 | 10.4% ±5 | Sep 2026 | Official leaderboard | |
| 29 | Claude Opus 5.5 | 9.17% ±4.6 | Sep 2026 | Official leaderboard | |
| 30 | | 8.75% ±4.3 | Aug 2026 | Official leaderboard | |
| 31 | | 7.50% ±4.4 | Oct 2026 | Official leaderboard | |
| 32 | GPT-6 Astra | 7.08% ±3.7 | Sep 2026 | Official leaderboard | |
| 33 | GPT-6.1 Sol | 6.25% ±3.6 | Sep 2026 | Official leaderboard | |
| 34 | Mercury 3 | 5.00% ±3.2 | Oct 2026 | Official leaderboard | |
| 35 | Claude Fable 5.2 | 5.00% ±3.4 | Nov 2026 | Official leaderboard | |
| 36 | Beam 1.0 | 4.58% ±3.5 | Oct 2026 | Official leaderboard | |
| 37 | GPT-6 Luna | 4.17% ±3.2 | Sep 2026 | Official leaderboard | |
| 38 | Claude Haiku 5.5 | 4.17% ±3 | Oct 2026 | Official leaderboard | |
| 39 | | 3.75% ±2.9 | Jul 2026 | Official leaderboard | |
| 40 | | 3.75% ±2.6 | Oct 2026 | Official leaderboard | |
| 41 | | 2.92% ±2.7 | Jul 2026 | Official leaderboard | |
| 42 | GPT-6.1 Terra | 1.67% ±2 | Oct 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | 27.1% | Sep 2026 | Official leaderboard | ||
| 2 | 25% | Oct 2026 | Official leaderboard | ||
| 3 | 25% | Nov 2026 | Official leaderboard | ||
| 4 | 23.8% | Oct 2026 | Official leaderboard | ||
| 5 | 23.3% | Oct 2026 | Official leaderboard | ||
| 6 | 21.7% | Nov 2026 | Lab model card | ||
| 7 | 21.3% | Nov 2026 | Lab model card | ||
| 8 | 19.6% | Nov 2026 | Official leaderboard | ||
| 9 | Grok 4.7 | 18.3% | Sep 2026 | Official leaderboard | |
| 10 | 17.5% | Oct 2026 | Official leaderboard | ||
| 11 | 17.5% | Oct 2026 | Official leaderboard | ||
| 12 | GPT-6.2 Astra | 16.3% | Nov 2026 | Official leaderboard | |
| 13 | 15.8% | Oct 2026 | Official leaderboard | ||
| 14 | 15.4% | Sep 2026 | Official leaderboard | ||
| 15 | GLM-5.3 | 14.6% | Aug 2026 | Official leaderboard | |
| 16 | 14.2% | Oct 2026 | Official leaderboard | ||
| 17 | MiMo-V2.6 Pro | 13.8% | Sep 2026 | Official leaderboard | |
| 18 | MiMo-V2.6 Flash | 13.3% | Sep 2026 | Official leaderboard | |
| 19 | Kimi K3 | 12.9% | Jul 2026 | Official leaderboard | |
| 20 | 12.9% | Sep 2026 | Official leaderboard | ||
| 21 | 12.5% | Nov 2026 | Official leaderboard | ||
| 22 | 12.1% | Sep 2026 | Official leaderboard | ||
| 23 | GPT-6.2 Sol | 12.1% | Nov 2026 | Official leaderboard | |
| 24 | Kimi K3.1 Pro | 11.7% | Oct 2026 | Official leaderboard | |
| 25 | 11.3% | Nov 2026 | Official leaderboard | ||
| 26 | 10.8% | Aug 2026 | Official leaderboard | ||
| 27 | GLM-5.3-Flash | 10.4% | Aug 2026 | Official leaderboard | |
| 28 | Claude Sonnet 5.5 | 10.4% | Sep 2026 | Official leaderboard | |
| 29 | Claude Opus 5.5 | 9.17% | Sep 2026 | Official leaderboard | |
| 30 | 8.75% | Aug 2026 | Official leaderboard | ||
| 31 | 7.50% | Oct 2026 | Official leaderboard | ||
| 32 | GPT-6 Astra | 7.08% | Sep 2026 | Official leaderboard | |
| 33 | GPT-6.1 Sol | 6.25% | Sep 2026 | Official leaderboard | |
| 34 | Mercury 3 | 5.00% | Oct 2026 | Official leaderboard | |
| 35 | Claude Fable 5.2 | 5.00% | Nov 2026 | Official leaderboard | |
| 36 | Beam 1.0 | 4.58% | Oct 2026 | Official leaderboard | |
| 37 | GPT-6 Luna | 4.17% | Sep 2026 | Official leaderboard | |
| 38 | Claude Haiku 5.5 | 4.17% | Oct 2026 | Official leaderboard | |
| 39 | 3.75% | Jul 2026 | Official leaderboard | ||
| 40 | 3.75% | Oct 2026 | Official leaderboard | ||
| 41 | 2.92% | Jul 2026 | Official leaderboard | ||
| 42 | GPT-6.1 Terra | 1.67% | Oct 2026 | Official leaderboard |
Labs report scores for 2 models in their cards. Switch the scores above to rank them together.
The best score, 27.1%, costs $11 a task
Runs on the cost frontier score higher than any run that's cheaper. Muse Ember comes within 2.1 points of the best for $0.12 a task, 88 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official Harvey LAB v1.1 (HLAB-Vals-1.1) leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| 27.1% | $11 | 153,535 | |||
| 25% | $14 | 115,480 | |||
| 25% | $0.12 | 95,364 | |||
| 23.8% | $7.08 | 100,926 | |||
| 23.3% | $5.41 | 112,495 | |||
| 21.7% | $1.23 | 170,530 | |||
| 21.3% | $13 | 447,456 | |||
| 19.6% | $5.97 | 175,645 | |||
| Grok 4.7 | 18.3% | $11 | 171,242 | ||
| 17.5% | $2.55 | 106,618 | |||
| 17.5% | $6.52 | 136,642 | |||
| GPT-6.2 Astra | 16.3% | $7.33 | 55,796 | ||
| 15.8% | $0.73 | 63,338 | |||
| 15.4% | $3.36 | 124,007 | |||
| GLM-5.3 | 14.6% | $3.99 | 135,019 | ||
| 14.2% | $0.61 | 215,683 | |||
| MiMo-V2.6 Pro | 13.8% | $0.22 | 86,713 | ||
| MiMo-V2.6 Flash | 13.3% | $0.09 | 103,528 | ||
| Kimi K3 | 12.9% | $4.14 | 83,561 | ||
| 12.9% | $1.13 | 118,439 | |||
| 12.5% | $1.04 | 100,731 | |||
| 12.1% | $3.66 | 137,535 | |||
| GPT-6.2 Sol | 12.1% | $2.25 | 67,160 | ||
| Kimi K3.1 Pro | 11.7% | $9.00 | 178,849 | ||
| 11.3% | $3.28 | 142,480 | |||
| 10.8% | $0.80 | 78,204 | |||
| GLM-5.3-Flash | 10.4% | $0.32 | 82,011 | ||
| Claude Sonnet 5.5 | 10.4% | $18 | 595,872 | ||
| Claude Opus 5.5 | 9.17% | $22 | 374,040 | ||
| 8.75% | $0.98 | 104,034 | |||
| 7.50% | $0.47 | 167,645 | |||
| GPT-6 Astra | 7.08% | $25 | 86,677 | ||
| GPT-6.1 Sol | 6.25% | $3.54 | 92,990 | ||
| Mercury 3 | 5.00% | $0.19 | 50,774 | ||
| Claude Fable 5.2 | 5.00% | $19 | 197,098 | ||
| Beam 1.0 | 4.58% | 36,042 | |||
| GPT-6 Luna | 4.17% | $0.30 | 103,034 | ||
| Claude Haiku 5.5 | 4.17% | $4.26 | 576,846 | ||
| 3.75% | $1.25 | 65,114 | |||
| 3.75% | $0.37 | 54,265 | |||
| 2.92% | $0.18 | 48,354 | |||
| GPT-6.1 Terra | 1.67% | $1.48 | 61,210 |
The best score, 27.1%, costs $11 a task
Runs on the cost frontier score higher than any run that's cheaper. Muse Ember comes within 2.1 points of the best for $0.12 a task, 88 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official Harvey LAB v1.1 (HLAB-Vals-1.1) leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| 27.1% | $11 | 153,535 | |||
| 25% | $14 | 115,480 | |||
| 25% | $0.12 | 95,364 | |||
| 23.8% | $7.08 | 100,926 | |||
| 23.3% | $5.41 | 112,495 | |||
| 21.7% | $1.23 | 170,530 | |||
| 21.3% | $13 | 447,456 | |||
| 19.6% | $5.97 | 175,645 | |||
| Grok 4.7 | 18.3% | $11 | 171,242 | ||
| 17.5% | $2.55 | 106,618 | |||
| 17.5% | $6.52 | 136,642 | |||
| GPT-6.2 Astra | 16.3% | $7.33 | 55,796 | ||
| 15.8% | $0.73 | 63,338 | |||
| 15.4% | $3.36 | 124,007 | |||
| GLM-5.3 | 14.6% | $3.99 | 135,019 | ||
| 14.2% | $0.61 | 215,683 | |||
| MiMo-V2.6 Pro | 13.8% | $0.22 | 86,713 | ||
| MiMo-V2.6 Flash | 13.3% | $0.09 | 103,528 | ||
| Kimi K3 | 12.9% | $4.14 | 83,561 | ||
| 12.9% | $1.13 | 118,439 | |||
| 12.5% | $1.04 | 100,731 | |||
| 12.1% | $3.66 | 137,535 | |||
| GPT-6.2 Sol | 12.1% | $2.25 | 67,160 | ||
| Kimi K3.1 Pro | 11.7% | $9.00 | 178,849 | ||
| 11.3% | $3.28 | 142,480 | |||
| 10.8% | $0.80 | 78,204 | |||
| GLM-5.3-Flash | 10.4% | $0.32 | 82,011 | ||
| Claude Sonnet 5.5 | 10.4% | $18 | 595,872 | ||
| Claude Opus 5.5 | 9.17% | $22 | 374,040 | ||
| 8.75% | $0.98 | 104,034 | |||
| 7.50% | $0.47 | 167,645 | |||
| GPT-6 Astra | 7.08% | $25 | 86,677 | ||
| GPT-6.1 Sol | 6.25% | $3.54 | 92,990 | ||
| Mercury 3 | 5.00% | $0.19 | 50,774 | ||
| Claude Fable 5.2 | 5.00% | $19 | 197,098 | ||
| Beam 1.0 | 4.58% | 36,042 | |||
| GPT-6 Luna | 4.17% | $0.30 | 103,034 | ||
| Claude Haiku 5.5 | 4.17% | $4.26 | 576,846 | ||
| 3.75% | $1.25 | 65,114 | |||
| 3.75% | $0.37 | 54,265 | |||
| 2.92% | $0.18 | 48,354 | |||
| GPT-6.1 Terra | 1.67% | $1.48 | 61,210 |
Number one over time Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Harvey LAB v1.1 (HLAB-Vals-1.1) leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Sources: Official Harvey LAB v1.1 (HLAB-Vals-1.1) leaderboard and labs’ model cards
2 labs put it in a launch post in the last six months
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| Nov 15, 2026 | 21.3 | Legal agent work, HLAB-Vals-1.1, cited from the Vals Index | |
| Nov 5, 2026 | 21.7 | Harvey LAB v1.1, 21.7% |
2 lab claims next to the leaderboard
Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. 2 of them match, which usually means one side copied the other. Every benchmark.
Versions
We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.
Source: Official leaderboards
More in Law, finance and medicine
- MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy.
Qwen 3.8 Max 80.4%
- HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
- HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
- FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy.
Ling-3.1-flash 81.7%
- KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
- Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%
Data audit
All 12 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access Public tasks, plus a private test set.
- Pass Confidence Checked against a primary source.
Scores 5 pass
- Pass Independent runs Scores from the official leaderboard (43 entries) and launch posts (2 scores).
- Pass Models scored 42 models we track, plus 1 other entry on its leaderboard.
- Pass Newest score The newest is from November 2026, for Gemini 4 Neon.
- Pass Frontier models 9 of the top 10 have a score, led by GPT-6.2 Astra (#1).
- Pass Headroom The best is 27.1%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on vals.ai: 42 rows for models we track and 1 other entry at the last read.
- Pass Last read November 19th, 2026.
- Pass Read worked The last read worked.
Consistency 2 pass
- Pass Lab claims Both pairs agree within 10 points.
- Pass Known issues No published audit found problems.