Harvey has unveiled Harvey Tenet, a new research preview model specifically designed for long-horizon legal tasks. This model is built upon the Kimi K3 base and enhanced through asynchronous reinforcement learning using Fireworks, with a training corpus that includes synthetic, public legal, and expert data. Harvey Tenet demonstrates significant improvements over its base model on legal benchmarks, achieving state-of-the-art results on LAB: Contracts and strong performance on the broader Harvey's Legal Agent Benchmark (LAB). Notably, these gains transfer to other agent platforms without retraining, indicating robust generalization capabilities. AI
IMPACT Sets new SOTA on legal benchmarks, potentially accelerating specialized AI adoption in the legal sector.
RANK_REASON Frontier-lab model release with system card. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
- Crosby
- Fireworks
- Harvey
- Harvey's Legal Agent Benchmark (LAB)
- Harvey Tenet
- Kimi k3
- LAB: Contracts
- LegalBench
- MAUD
- Mercor
- Nvidia B300 GPUs
- PRBench
- Redline Bench
- Scale
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
