Moonshot AIPremium plan
Kimi K2.7 Code Highspeed
Moonshot AI's Kimi K2.7 Code Highspeed via the Vercel AI Gateway: a low-latency, high-throughput serving tier of Kimi K2.7 Code for long-horizon coding, coding-driven design, autonomous execution, and task orchestration. 262k context. Supports image input and function calling. Uses 3 premium requests per send before length multipliers. Does not support web search.
You always get the exact model you pick — we never silently route you to another.
Specifications
| Provider | Moonshot AI |
|---|---|
| Released | 2026-06 |
| Intelligence | High |
| Speed | Fast |
| Context window | 262,144 tokens |
| Max output | 262,144 tokens |
| Knowledge cutoff | Not publicly disclosed |
| Input price | $1.90 / 1M tokens |
| Output price | $8.00 / 1M tokens |
| Request cost | 3 premium requests |
| Plan tier | Premium |
| Input | Text, Image |
| Output | Text |
| Features | Cached input: $0.38 / 1M tokens, 3 premium requests per send before length multipliers, Function calling supported, Served through the Vercel AI Gateway (moonshotai/kimi-k2.7-code-highspeed) |
| Model ID | kimi-k2p7-code-highspeed |