Kimi K2.7 Code

Moonshot AI🇨🇳 China
active
Context window256K tokens
Input / 1M tokens$0.82
Output / 1M tokens$3.75

Version History

2.7minor

Kimi K2.7 Code builds on K2.6 with substantial improvements in real-world long-horizon coding tasks and 30% reduction in thinking token usage. Added experimental video input support and improved performance across coding and agentic benchmarks.

Benchmark Scores

Full leaderboard →
89.6%
GPQA
80.3%
Hallucination Rate
45.0 tokens_per_sec
Speed (tok/s)

Coverage