Kimi K2.7 Code is a coding-focused agentic model built upon Kimi K2.6. With substantial improvements on real-world long-horizon coding tasks, it strengthens end-to-end task completion across complex software engineering workflows while improving token efficiency, reducing thinking-token usage by approximately 30% compared with Kimi K2.6.
Key Features
- Architecture: Mixture-of-Experts (MoE) with 1T total parameters, 32B activated
- Context Length: 256K tokens
- Vision Encoder: MoonViT with 400M parameters
- Native INT4 Quantization: Same method as Kimi-K2-Thinking
- Deployment: Supports vLLM, SGLang, KTransformers
- License: Modified MIT
Use Cases
- Complex software engineering tasks
- Coding agent scenarios with multi-turn interactions
- Image and video input processing
- Interleaved thinking and multi-step tool calls
- Best used with Kimi Code CLI agent framework
