Kimi Releases K3.1 Pro, Says It Beats GPT-6 Astra on Coding Test
Summary
Kimi releases K3.1 Pro, a 3T-class open-weight model with a 1-million-token context window, on Kimi.ai, Kimi Work, Kimi Code and the Kimi API. The company says it scores 65.7% on Terminal-Bench 4.0, ahead of GPT-6 Astra at 58.2%. A Highspeed variant runs 2-3× faster. Weights are due by Nov 4.
Key Points
- Kimi says Claude Opus 5.5 edges K3.1 Pro on Terminal-Bench 4.0 at 66.4%, against K3.1 Pro's 65.7%.
- Kimi Enterprise administrators can enable K3.1 Pro for their organization, but access is off by default at launch.
- Kimi reports the claim-level hallucination rate on its Kimi Hallu Bench Hard test falls to 8.3% from K3's 23.4%.