Harvey Introduces Harvey Tenet: A Kimi K3 Base Post-Trained with Fireworks for Long-Horizon Legal Agent Work

Harvey introduced Harvey Tenet, a research-preview legal AI model post-trained from Kimi K3 for long-horizon legal work.

Trained with asynchronous reinforcement learning on synthetic, public legal, and expert-labeled data, Tenet outperformed its base model on benchmarks for contract review and M&A diligence.

Weights and an API are not yet available, but the effort points to how law firms and legal teams could build or own specialized agentic models without sacrificing general legal reasoning.