跳到主要内容
非官方简体中文翻译，内容版权归 Mastra 所有，遵循 Apache-2.0 许可。

Kenari logoKenari

Access 64 Kenari models through Mastra's model router. Authentication is handled automatically using the KENARI_API_KEY environment variable.

Learn more in the Kenari documentation.

.env
KENARI_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: "kenari/claude-fable-5"
});

// 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);
}
备注

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

Models
Models的直接链接

ModelContextToolsReasoningImageAudioVideoInput $/1MOutput $/1M
kenari/claude-fable-51.0M——
kenari/claude-opus-4-71.0M——
kenari/claude-opus-4-81.0M——
kenari/claude-opus-51.0M——
kenari/claude-sonnet-4-61.0M——
kenari/claude-sonnet-51.0M——
kenari/deepseek-v4-1-flash1.0M——
kenari/deepseek-v4-flash1.0M——
kenari/deepseek-v4-flash:free1.0M——
kenari/deepseek-v4-pro1.0M——
kenari/gemini-2-5-flash1.0M——
kenari/gemini-2-5-flash-lite1.0M——
kenari/gemini-3-1-flash-lite1.0M——
kenari/gemini-3-1-flash-tts8K——
kenari/gemini-3-1-pro1.0M——
kenari/gemini-3-5-flash1.0M——
kenari/gemini-3-6-flash1.0M——
kenari/gemini-3-7-flash1.0M——
kenari/gemma-4-31b-it262K——
kenari/glm-4-7-flash:free200K——
kenari/glm-5-1200K——
kenari/glm-5-21.0M——
kenari/glm-5-31.0M——
kenari/glm-5-3-flash1.0M——
kenari/gpt-5-4-mini400K——
kenari/gpt-5-51.1M——
kenari/gpt-5-6-luna1.1M——
kenari/gpt-5-6-sol1.1M——
kenari/gpt-5-6-terra1.1M——
kenari/gpt-image-2272K——
kenari/gpt-oss-120b131K——
kenari/gpt-oss-20b131K——
kenari/grok-4-5500K——
kenari/grok-4-6500K——
kenari/grok-build-0-1256K——
kenari/grok-imagine-image16K——
kenari/grok-imagine-image-2-064K——
kenari/hy3256K——
kenari/hy3:free256K——
kenari/kimi-k2-6262K——
kenari/kimi-k2-6:free262K——
kenari/kimi-k2-7-code262K——
kenari/kimi-k2-7-code:free262K——
kenari/kimi-k31.0M——
kenari/mimo-v2-51.0M——
kenari/mimo-v2-5-pro1.0M——
kenari/mimo-v2-5:free1.0M——
kenari/minimax-m2-7205K——
kenari/minimax-m2-7-highspeed205K——
kenari/minimax-m31.0M——
kenari/mistral-large:free262K——
kenari/mistral-medium-3-5:free262K——
kenari/nano-banana-2131K——
kenari/nano-banana-2-lite66K——
kenari/nano-banana-pro66K——
kenari/nemotron-3-nano-30b-a3b262K——
kenari/nemotron-3-super-120b-a12b262K——
kenari/nemotron-3-super-120b-a12b:free262K——
kenari/nemotron-3-ultra-550b-a55b1.0M——
kenari/qwen3-7-plus1.0M——
kenari/qwen3-8-max1.0M——
kenari/step-3-7-flash256K——
kenari/step-3-7-flash:free256K——
kenari/whisper-large-v3-turbo448——
64 available models

Model availability, capabilities, context windows, and pricing are sourced from models.dev and may change.

Advanced configuration
Advanced configuration的直接链接

Custom headers
Custom headers的直接链接

src/mastra/agents/my-agent.ts
const agent = new Agent({
  id: "custom-agent",
  name: "custom-agent",
  model: {
    url: "https://kenari.id/v1",
    id: "kenari/claude-fable-5",
    apiKey: process.env.KENARI_API_KEY,
    headers: {
      "X-Custom-Header": "value"
    }
  }
});

Dynamic model selection
Dynamic model selection的直接链接

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
      ? "kenari/whisper-large-v3-turbo"
      : "kenari/claude-fable-5";
  }
});
On this page