Atomic Chat

Access 5 Atomic Chat models through Mastra's model router. Authentication is handled automatically using the ATOMIC_CHAT_API_KEY environment variable.

Learn more in the Atomic Chat documentation.

.env

ATOMIC_CHAT_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: "atomic-chat/Meta-Llama-3_1-8B-Instruct-GGUF"
});

// 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 Atomic Chat documentation for details.

Models

5 available models

Model	Context	Tools	Reasoning	Image	Audio	Video	Input $/1M	Output $/1M
atomic-chat/gemma-4-E4B-it-IQ4_XS	33K						—	—
atomic-chat/gemma-4-E4B-it-MLX-4bit	33K						—	—
atomic-chat/Meta-Llama-3_1-8B-Instruct-GGUF	131K						—	—
atomic-chat/Qwen3_5-9B-MLX-4bit	33K						—	—
atomic-chat/Qwen3_5-9B-Q4_K_M	33K						—	—

Advanced configuration

Custom headers

src/mastra/agents/my-agent.ts

const agent = new Agent({
  id: "custom-agent",
  name: "custom-agent",
  model: {
    url: "http://127.0.0.1:1337/v1",
    id: "atomic-chat/Meta-Llama-3_1-8B-Instruct-GGUF",
    apiKey: process.env.ATOMIC_CHAT_API_KEY,
    headers: {
      "X-Custom-Header": "value"
    }
  }
});

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
      ? "atomic-chat/gemma-4-E4B-it-MLX-4bit"
      : "atomic-chat/Meta-Llama-3_1-8B-Instruct-GGUF";
  }
});