Skip to main content
Evaluation sources/Harvey Legal Agent Benchmark
ParticipatingProfessional work

Harvey Legal Agent Benchmark

Vals AI / Harvey · Public model evaluation results from Harvey Legal Agent Benchmark.

Official evaluation
Source description

This source measures concrete delivery performance in Professional work; raw scores are shown separately from the normalized capability score.

Limitation: Different versions, reasoning settings, and test sets may change scores; read them with the data date.

How it is used

Participating sources affect capability scores; reference, cross-reference, and observing sources are for comparison and verification only.

Model results from this source

Gemini 4 Argon27.0892.92026-11-05
Muse Spark 1.225.4288.62026-10-18
Muse Spark 1.323.7584.22026-10-18
Muse Spark 1.12074.52026-10-18
Grok 4.718.3370.12026-11-05
Grok 4.615.8363.62026-10-18
Mistral Large 415.8363.62026-11-05
GLM 5.314.5860.42026-11-05
DeepSeek V4.1 Flash 101714.1759.32026-11-05
MiMo V2.6 Pro13.7558.22026-11-05
MiMo V2.6 Flash13.3357.12026-11-05
Grok 4.512.9256.02026-10-18
Kimi K312.9256.02026-11-05
Step 5 Preview12.9256.02026-11-05
Gemini 3.8 Flash12.0853.82026-11-05
MiMo V2.6 Flash11.2551.72026-10-18
MiMo V2.6 Flash11.2551.72026-10-17
Claude Fable 511.2551.72026-10-18
DeepSeek V4 Pro 081310.8350.62026-11-05
Qwen3.8 Max10.4249.52026-10-18
GLM-5.3 Flash10.4249.52026-11-05
Claude Sonnet 5.510.4249.52026-11-05
Claude Opus 4.89.5847.32026-10-18
Claude Opus 5.59.1746.32026-11-05
DeepSeek V4 Flash 07318.3344.12026-10-18
MiniMax M3.1 Flash7.541.92026-11-05
GLM-5.27.0840.82026-10-18
GPT-6 Astra7.0840.82026-11-05
Claude Opus 4.76.6739.72026-10-18
Claude Opus 56.6739.72026-10-18
GLM-5.3 Flash6.6739.72026-10-18
GLM-5.3 Flash6.6739.72026-10-17
Claude Fable 5.16.6739.72026-10-18
DeepSeek V4.1 Flash6.6739.72026-10-18
GPT-6.1 Sol6.2538.62026-11-05
Claude Sonnet 4.6535.42026-10-18
Claude Sonnet 5535.42026-10-18
MiniMax-M34.1733.22026-10-18
GPT-6 Luna4.1733.22026-11-05
Claude Haiku 5.54.1733.22026-11-05
GPT-5.53.7532.12026-10-18
Inkling3.7532.12026-11-05
Gemini 3.6 Flash3.3331.02026-10-18
Inkling Small2.9230.02026-11-05
GPT-5.6 Sol2.528.92026-10-18
MiMo-V2.5-Pro2.0827.82026-10-18
Kimi K2.61.6726.72026-10-18
MiMo-V2.51.6726.72026-10-18
Qwen3.7 Max1.6726.72026-10-18
GPT-6 Sol1.6726.72026-10-18
GPT-6.1 Terra1.6726.72026-11-05
Qwen3.6 Plus1.2525.62026-10-18
GPT-5.6 Luna1.2525.62026-10-18
Ling-3.0-flash (free)1.2525.62026-10-18
Claude Haiku 4.50.8324.52026-10-18
GPT-5.6 Terra0.8324.52026-10-18
Grok 4.30.4223.52026-10-18
Mistral Medium 3.50.4223.52026-10-11
Kimi K2.5022.42026-10-18
Gemini 3.1 Flash Lite Preview022.42026-10-18
GPT-5.4022.42026-10-18
Grok 4.20 (Reasoning)022.42026-10-18
GPT-5.4 mini022.42026-10-18
MiniMax-M2.7022.42026-10-18
GLM-5.1022.42026-10-18
Laguna M.1022.42026-10-18
Laguna XS.2022.42026-10-18
Qwen3.7 Plus022.42026-10-18
Gemini 3.5 Flash Lite022.42026-10-18

Frequently asked questions