Skip to main content

Qiniu logoQiniu

Access 76 Qiniu models through Mastra's model router. Authentication is handled automatically using the Qiniu_API_KEY environment variable.

Learn more in the Qiniu documentation.

.env
Qiniu_API_KEY=your-api-key
src/mastra/agents/my-agent.ts
import { Agent } from "@mastra/core/agent";

const agent = new Agent({
  id: "my-agent",
  name: "My Agent",
  instructions: "You are a helpful assistant",
  model: "qiniu-ai/MiniMax-M1"
});

// Generate a response
const response = await agent.generate("Hello!");

// Stream a response
const stream = await agent.stream("Tell me a story");
for await (const chunk of stream) {
  console.log(chunk);
}
info

Mastra uses the OpenAI-compatible /chat/completions endpoint. Some provider-specific features may not be available. Check the Qiniu documentation for details.

Models
Direct link to Models

ModelContextToolsReasoningImageAudioVideoInput $/1MOutput $/1M
qiniu-ai/claude-3.5-haiku200K——
qiniu-ai/claude-3.5-sonnet200K——
qiniu-ai/claude-3.7-sonnet200K——
qiniu-ai/claude-4.0-opus200K——
qiniu-ai/claude-4.0-sonnet200K——
qiniu-ai/claude-4.1-opus200K——
qiniu-ai/claude-4.5-haiku200K——
qiniu-ai/claude-4.5-opus200K——
qiniu-ai/claude-4.5-sonnet200K——
qiniu-ai/deepseek-r1128K——
qiniu-ai/deepseek-r1-0528128K——
qiniu-ai/deepseek-v3128K——
qiniu-ai/deepseek-v3-0324128K——
qiniu-ai/deepseek-v3.1128K——
qiniu-ai/deepseek/deepseek-math-v2160K——
qiniu-ai/deepseek/deepseek-v3.1-terminus128K——
qiniu-ai/deepseek/deepseek-v3.1-terminus-thinking128K——
qiniu-ai/deepseek/deepseek-v3.2-251201128K——
qiniu-ai/deepseek/deepseek-v3.2-exp128K——
qiniu-ai/deepseek/deepseek-v3.2-exp-thinking128K——
qiniu-ai/doubao-1.5-pro-32k128K——
qiniu-ai/doubao-1.5-thinking-pro128K——
qiniu-ai/doubao-1.5-vision-pro128K——
qiniu-ai/doubao-seed-1.6256K——
qiniu-ai/doubao-seed-1.6-flash256K——
qiniu-ai/doubao-seed-1.6-thinking256K——
qiniu-ai/gemini-2.0-flash1.0M——
qiniu-ai/gemini-2.0-flash-lite1.0M——
qiniu-ai/gemini-2.5-flash1.0M——
qiniu-ai/gemini-2.5-flash-image33K——
qiniu-ai/gemini-2.5-flash-lite1.0M——
qiniu-ai/gemini-2.5-pro1.0M——
qiniu-ai/gemini-3.0-flash-preview1.0M——
qiniu-ai/gemini-3.0-pro-image-preview33K——
qiniu-ai/gemini-3.0-pro-preview1.0M——
qiniu-ai/glm-4.5131K——
qiniu-ai/glm-4.5-air131K——
qiniu-ai/gpt-oss-120b128K——
qiniu-ai/gpt-oss-20b128K——
qiniu-ai/kimi-k2128K——
qiniu-ai/kling-v2-6100.0M——
qiniu-ai/meituan/longcat-flash-chat131K——
qiniu-ai/mimo-v2-flash256K——
qiniu-ai/MiniMax-M11.0M——
qiniu-ai/minimax/minimax-m2200K——
qiniu-ai/minimax/minimax-m2.1205K——
qiniu-ai/moonshotai/kimi-k2-0905256K——
qiniu-ai/moonshotai/kimi-k2-thinking256K——
qiniu-ai/openai/gpt-5400K——
qiniu-ai/openai/gpt-5.2400K——
qiniu-ai/qwen-max-2025-01-25128K——
qiniu-ai/qwen-turbo1.0M——
qiniu-ai/qwen-vl-max-2025-01-25128K——
qiniu-ai/qwen2.5-vl-72b-instruct128K——
qiniu-ai/qwen2.5-vl-7b-instruct128K——
qiniu-ai/qwen3-235b-a22b128K——
qiniu-ai/qwen3-235b-a22b-instruct-2507262K——
qiniu-ai/qwen3-235b-a22b-thinking-2507262K——
qiniu-ai/qwen3-30b-a3b40K——
qiniu-ai/qwen3-32b40K——
qiniu-ai/qwen3-coder-480b-a35b-instruct262K——
qiniu-ai/qwen3-max262K——
qiniu-ai/qwen3-max-preview256K——
qiniu-ai/qwen3-next-80b-a3b-instruct131K——
qiniu-ai/qwen3-next-80b-a3b-thinking131K——
qiniu-ai/stepfun-ai/gelab-zero-4b-preview8K——
qiniu-ai/x-ai/grok-4-fast2.0M——
qiniu-ai/x-ai/grok-4-fast-non-reasoning2.0M——
qiniu-ai/x-ai/grok-4-fast-reasoning2.0M——
qiniu-ai/x-ai/grok-4.1-fast2.0M——
qiniu-ai/x-ai/grok-4.1-fast-non-reasoning2.0M——
qiniu-ai/x-ai/grok-4.1-fast-reasoning20.0M——
qiniu-ai/x-ai/grok-code-fast-1256K——
qiniu-ai/z-ai/autoglm-phone-9b13K——
qiniu-ai/z-ai/glm-4.6200K——
qiniu-ai/z-ai/glm-4.7200K——
76 available models

Advanced Configuration
Direct link to Advanced Configuration

Custom Headers
Direct link to Custom Headers

src/mastra/agents/my-agent.ts
const agent = new Agent({
  id: "custom-agent",
  name: "custom-agent",
  model: {
    url: "https://api.qnaigc.com.com/v1",
    id: "qiniu-ai/MiniMax-M1",
    apiKey: process.env.Qiniu_API_KEY,
    headers: {
      "X-Custom-Header": "value"
    }
  }
});

Dynamic Model Selection
Direct link to Dynamic Model Selection

src/mastra/agents/my-agent.ts
const agent = new Agent({
  id: "dynamic-agent",
  name: "Dynamic Agent",
  model: ({ requestContext }) => {
    const useAdvanced = requestContext.task === "complex";
    return useAdvanced
      ? "qiniu-ai/z-ai/glm-4.7"
      : "qiniu-ai/MiniMax-M1";
  }
});
On this page