Publish time
2026/4/16Model Series
千问Input type
Output type
Input Price
¥1.2 / 1M tokensOutput Price
¥7.2 / 1M tokensCache Write Price
¥1.5 / 1M tokensCache Read Price
¥0.12 / 1M tokensContext Window
991,000Max Output Length
64,000Qwen3.6原生视觉语言系列Flash模型,模型效果相较3.5-Flash显著提升。本模型重点提升agentic coding能力(在多项代码智能体基准上大幅超越前代)、数学推理和代码推理能力;视觉方面在空间智能能力上显著增强,物体定位与目标检测提升尤为突出。
Zhinao API routes requests to the best-fit provider and automatically fails over to the one with highest availability.
TTFT
0.63s
Throughput
114.01tps
Uptime
100.00%
Provider Model
alibaba/qwen/qwen3.6-flash
Supported Parameters
Recent Uptime
Reasoning
Toggleable
Supported Response Formats
Total Context
991,000
Max Output
64,000
Input Price
¥1.2 / 1M tokens
Output Price
¥7.2 / 1M tokens
Cache Write
¥1.5 / 1M tokens
Cache Read
¥0.12 / 1M tokens
Compare different providers across Zhinao API
114.01 tok/s
0.63 s
Uptime for qwen/qwen3.6-flash 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: "qwen/qwen3.6-flash",
messages: [
{ role: "user", content: "Hello, how are you?" }
],
temperature: 0.7,
max_tokens: 1000,
});
console.log(response.choices[0].message.content);