Kimi K2.7 Code
Dedicated coding model; pairs with Kimi Code CLI. Release day not verified (HF 2026-06-11, OpenRouter 2026-06-12). Max output not verified.
- Context window
- 262,144 tokens
- Input
- text, image, video
- Output
- text
- License
- modified-mit
- Pricing
- input: $0.95 · output: $4 · cache read: $0.19 (per 1M tokens (USD); kimi-k2.7-code-highspeed: 1.90 in / 8.00 out / 0.38 cache hit) source
- Verified
- 2026-09-29
How to call it
| Provider | Model id | Endpoint / URL | Docs |
|---|---|---|---|
| Kimi API (Moonshot) | kimi-k2.7-code | https://api.moonshot.ai/v1 | docs |
| Kimi API (Moonshot) high-speed | kimi-k2.7-code-highspeed | https://api.moonshot.ai/v1 | docs |
| OpenRouter | moonshotai/kimi-k2.7-code | openrouter.ai/moonshotai/kimi-k2.7-code | — |
| Hugging Face | — | huggingface.co/moonshotai/Kimi-K2.7-Code | — |
| Web app | — | www.kimi.com | — |
Notable capabilities (3)
- Coding-specialized K2.6 derivative: Built on Kimi K2.6 (1T total / 32B active, MLA, 400M vision encoder) and tuned for long-horizon real-world coding. source
- ~30% fewer thinking tokens than K2.6: Higher task success with about 30% lower thinking-token usage vs K2.6. source
- High-speed tier: kimi-k2.7-code-highspeed outputs ~180 tok/s (up to ~260 tok/s on short context). source
Cheaper-than-K3 coding agent model for Kimi Code CLI, Claude Code, Codex, OpenCode.
curl https://api.moonshot.ai/v1/chat/completions \
-H "Authorization: Bearer $MOONSHOT_API_KEY" -H "Content-Type: application/json" \
-d '{"model":"kimi-k2.7-code","messages":[{"role":"user","content":"Write a Python quicksort"}]}'
Sources: https://platform.kimi.ai/docs/models · https://platform.kimi.ai/docs/pricing/chat · https://huggingface.co/moonshotai/Kimi-K2.7-Code