agentic
Harvey's Legal Agent Benchmark
- #39max of 793.8%
Leader: Muse Spark 1.2 · 25.4%
DeepSeek
Measured on 28 benchmarks.
General frontier
Not ranked in this mix — too little measured data for its benchmarks.
This model has no results on your mix’s benchmarks.
agentic
Leader: Muse Spark 1.2 · 25.4%
agentic
Leader: Muse Spark 1.3 · 55.3%
agentic
Leader: DeepSeek V4.1 Flash · 69.8%
agentic
Leader: Fable 5.1 · 49.2%
agentic
Leader: Opus 5.5 · 65.2%
coding
Leader: Sonnet 5.5 · 69.8%
coding
Leader: Fable 5.1 · 90.5%
coding
Leader: Opus 5.5 · 65.0%
coding
Leader: Opus 5.5 · 18.5%
coding
Leader: Opus 5.5 · 87.0%
coding
Leader: Opus 5 · 97.0%
coding
Leader: GPT-5.5 · 73.2%
coding
Leader: Opus 5.5 · 87.6%
coding
Leader: Sonnet 5.5 · 92.4%
composite
Leader: Gemini 4 · 68.9%
economics
Leader: Opus 5 · 73.2%
economics
Leader: Fable 5.1 · 76.7%
economics
Leader: Gemini 4 · 65.4%
economics
Leader: Fable 5 · 71.7%
economics
Leader: Muse Spark 1.2 · 80.4%
knowledge
Leader: Grok 4.3 · 79.3%
knowledge
Leader: Gemini 3.1 Pro · 95.5%
knowledge
Leader: Opus 5 · 63.6%
knowledge
Leader: Opus 5.5 · 91.4%
knowledge
Leader: Fable 5.1 · 92.4%
knowledge
Leader: Opus 5 · 76.9%
math
Leader: Sonnet 5.5 · 100.0%
reasoning
Leader: Fable 5 · 88.6%