Publish time
2026/6/30Model Series
KimiInput type
Output type
Input Price
¥13 / 1M tokensOutput Price
¥54 / 1M tokensCache Write Price
¥16.25 / 1M tokensCache Read Price
¥1.3 / 1M tokensContext Window
262,144Max Output Length
262,144Kimi K2.7 Code 高速版，与 Kimi K2.7 Code 是同一个模型，但输出速度约为普通版的 5-6 倍，常规编程场景下（取输入长度中位数）输出速度约 180 Token/s，短上下文场景可达 260 Token/s ，支持 256K 上下文窗口，带来更极致的编程体验。
Zhinao API routes requests to the best-fit provider and automatically fails over to the one with highest availability.
TTFT
6.05s
Throughput
62.24tps
Uptime
100.00%
Provider Model
moonshot/moonshotai/kimi-k2.7-code-highspeed
Supported Parameters
Recent Uptime
Reasoning
Toggleable
Supported Response Formats
Total Context
262,144
Max Output
262,144
Input Price
¥13 / 1M tokens
Output Price
¥54 / 1M tokens
Cache Write
¥16.25 / 1M tokens
Cache Read
¥2.6 / 1M tokens
Compare different providers across Zhinao API
61.90 tok/s
6.05 s
Uptime for moonshotai/kimi-k2.7-code-highspeed across all providers
Zhinao API normalizes requests and responses across providers for you
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.360.cn/v1",
apiKey: process.env.ZHINAO_API_KEY,
});
const response = await client.chat.completions.create({
model: "moonshotai/kimi-k2.7-code-highspeed",
messages: [
{ role: "user", content: "Hello, how are you?" }
],
temperature: 0.7,
max_tokens: 1000,
});
console.log(response.choices[0].message.content);