Moonshot's current coding model for software-engineering agents with image/video input and tool use.
Moonshot's quickstart describes K2.7 Code as an always-thinking tool-use model, which can increase end-to-end latency and output cost in long agent loops.
ActionMeasure completion time and total tokens on real repository tasks, not only single-response speed.
Verify source| Company | Moonshot AI / Kimi |
|---|---|
| Release | 2026-06-12 |
| Release type | Model release |
| Parameters | Not disclosed |
| Architecture | Not disclosed |
| Context | 262,144 |
| Input modalities | text / image / video |
| Output modalities | text / code / tool_call |
| Tool use / function calling | Confirmed: Tool use, Function calling |
| Structured output / reasoning | Confirmed: Reasoning |
| Vision / audio / video | Confirmed: Vision, Video |
| Open weights / License | Weights available / Modified MIT (license source pending verification) moonshot-kimi-kimi-k2-7-code-current · License source pending verification |
| Vendor API | Available Kimi Code what's newmoonshot-kimi-kimi-k2-7-code-current · 2026-06-12 |
| OpenRouter | OpenRouter listed |
| Pricing | OpenRouter · 2026-07-15: $0.719 / $3.49 input / output, per 1M tokens |
| Benchmark | No reliable public source |
| Public signals | No reliable public source |
| Use cases | Software engineering agents / Repository-scale coding / Visual coding tasks / local inference |
| Claim evidence | Release eventKimi Code what's newkimi-k2.7-code · 2026-06-12 Model specs / weightsKimi K2.7 Code quickstartKimi K2.7 Code · kimi-k2.7-code |
| OpenRouter | OpenRouter listed |
|---|---|
| OpenRouter listing | moonshotai/kimi-k2.7-code |
| Mapped version | kimi-k2.7-code moonshot-kimi-kimi-k2-7-code-current |
| Pricing | OpenRouter · 2026-07-15: $0.719 / $3.49 input / output, per 1M tokens |
| Context | 262144 |
| Speed/Latency | No reliable public source |
| Snapshot time | 2026-07-15T20:11:41.551Z |
| Source | View source |