AI EVALUATION, IN CONTEXT.BY EQUALITY MACHINE
Benchmark Radar.Benchmarks. Model results. Original sources.
/

Artificial Analysis

Harvey LAB-AA

Legal agentic work, criterion pass rate

Evaluation context and evidence

Source
Artificial Analysis
Publisher
Not recorded
Benchmark release
Not recorded
Collected
2026-08-25
Modality
Not recorded

These are source-reported observations, not independent reruns by this site. Check the task version, evaluation protocol, unit and reporting date before comparing models. Results from different setups are not a universal ranking.

Original sources

Reported results 36

Shown in source observation order, not ranked. Dates retain the precision provided by the source.

ModelValueUnitProtocolReported date and basisSource
Kimi K3 (max)0.946446663771892Not recordedNot recorded2026-07-16 · model_announcementArtificial Analysis
Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)0.935591257779708Not recordedNot recorded2026-06-09 · model_announcementArtificial Analysis
Grok 4.5 (high)0.924156896801274Not recordedNot recorded2026-07-08 · model_announcementArtificial Analysis
Claude Opus 4.8 (Adaptive Reasoning, Max Effort)0.910840932117528Not recordedNot recorded2026-05-28 · model_announcementArtificial Analysis
GLM-5.2 (max)0.909683022145028Not recordedNot recorded2026-06-16 · model_announcementArtificial Analysis
Gemini 3.7 Flash (high)0.906643508467217Not recordedNot recorded2026-08-13 · model_announcementArtificial Analysis
Claude Sonnet 5 (Adaptive Reasoning, Max Effort)0.900709219858156Not recordedNot recorded2026-06-30 · model_announcementArtificial Analysis
MiniMax-M30.884064264003474Not recordedNot recorded2026-06-01 · model_announcementArtificial Analysis
GPT-5.6 Luna (max)0.878998407873788Not recordedNot recorded2026-07-09 · model_announcementArtificial Analysis
GPT-5.6 Sol (max)0.871761470545665Not recordedNot recorded2026-07-09 · model_announcementArtificial Analysis
GPT-5.5 (xhigh)0.862787668258793Not recordedNot recorded2026-04-23 · model_announcementArtificial Analysis
Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)0.859748154580981Not recordedNot recorded2026-02-17 · model_announcementArtificial Analysis
GPT-5.6 Terra (max)0.851787523520046Not recordedNot recorded2026-07-09 · model_announcementArtificial Analysis
Gemini 3.6 Flash (high)0.851498046026921Not recordedNot recorded2026-07-21 · model_announcementArtificial Analysis
Kimi K2.7 Code0.850195397307859Not recordedNot recorded2026-06-12 · model_announcementArtificial Analysis
Claude Sonnet 4.6 (Non-reasoning, High Effort)0.844405847445361Not recordedNot recorded2026-02-17 · model_announcementArtificial Analysis
DeepSeek V4 Pro (Reasoning, Max Effort)0.843971631205674Not recordedNot recorded2026-04-24 · model_announcementArtificial Analysis
Kimi K2.60.841221595020987Not recordedNot recorded2026-04-20 · model_announcementArtificial Analysis
Qwen3.7 Max0.834274135185989Not recordedNot recorded2026-05-19 · model_announcementArtificial Analysis
Qwen3.6 27B (Reasoning)0.822984512954118Not recordedNot recorded2026-04-22 · model_announcementArtificial Analysis
Gemini 3.5 Flash (high)0.820813431755681Not recordedNot recorded2026-05-19 · model_announcementArtificial Analysis
Qwen3.7 Plus0.817918656824432Not recordedNot recorded2026-06-01 · model_announcementArtificial Analysis
Nemotron 3 Ultra 550B A55B (Reasoning)0.81719496309162Not recordedNot recorded2026-06-04 · model_announcementArtificial Analysis
DeepSeek V4 Flash (Reasoning, Max Effort)0.813287016934433Not recordedNot recorded2026-04-24 · model_announcementArtificial Analysis
MiMo-V2.5-Pro0.732522796352584Not recordedNot recorded2026-04-22 · model_announcementArtificial Analysis
Step 3.7 Flash0.727312201476335Not recordedNot recorded2026-05-29 · model_announcementArtificial Analysis
Qwen3.5 397B A17B (Reasoning)0.723838471558836Not recordedNot recorded2026-02-16 · model_announcementArtificial Analysis
Mistral Medium 3.50.690982776089159Not recordedNot recorded2026-04-29 · model_announcementArtificial Analysis
Grok 4.3 (high)0.683890577507599Not recordedNot recorded2026-04-30 · model_announcementArtificial Analysis
Claude 4.5 Haiku (Reasoning)0.610508033000434Not recordedNot recorded2025-10-15 · model_announcementArtificial Analysis
GPT-5.4 mini (xhigh)0.607468519322623Not recordedNot recorded2026-03-17 · model_announcementArtificial Analysis
Gemini 3.1 Pro Preview0.588507743522941Not recordedNot recorded2026-02-19 · model_announcementArtificial Analysis
GPT-5.4 nano (xhigh)0.522362136343899Not recordedNot recorded2026-03-17 · model_announcementArtificial Analysis
Gemma 4 31B (Reasoning)0.47228253003329Not recordedNot recorded2026-04-02 · model_announcementArtificial Analysis
Gemini 3.1 Flash-Lite0.311477782602403Not recordedNot recorded2026-03-03 · model_announcementArtificial Analysis
gpt-oss-120b (high)0.138949196699957Not recordedNot recorded2025-08-05 · model_announcementArtificial Analysis

Put this benchmark in context

Tools and agents benchmarks