Hugging Face

Access 15 Hugging Face models through Mastra's model router. Authentication is handled automatically using the HF_TOKEN environment variable.

Learn more in the Hugging Face documentation.

HF_TOKEN=your-api-key

import { Agent } from "@mastra/core";

const agent = new Agent({
  name: "my-agent",
  instructions: "You are a helpful assistant",
  model: "huggingface/MiniMaxAI/MiniMax-M2.1"
});

// 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 Hugging Face documentation for details.

Models

15 available models

Model	Context	Tools	Reasoning	Image	Audio	Video	Input $/1M	Output $/1M
huggingface/deepseek-ai/DeepSeek-R1-0528	164K						$3	$5
huggingface/deepseek-ai/DeepSeek-V3.2	164K						$0.28	$0.40
huggingface/MiniMaxAI/MiniMax-M2.1	205K						$0.30	$1
huggingface/moonshotai/Kimi-K2-Instruct	131K						$1	$3
huggingface/moonshotai/Kimi-K2-Instruct-0905	262K						$1	$3
huggingface/moonshotai/Kimi-K2-Thinking	262K						$0.60	$3
huggingface/Qwen/Qwen3-235B-A22B-Thinking-2507	262K						$0.30	$3
huggingface/Qwen/Qwen3-Coder-480B-A35B-Instruct	262K						$2	$2
huggingface/Qwen/Qwen3-Embedding-4B	32K						$0.01	—
huggingface/Qwen/Qwen3-Embedding-8B	32K						$0.01	—
huggingface/Qwen/Qwen3-Next-80B-A3B-Instruct	262K						$0.25	$1
huggingface/Qwen/Qwen3-Next-80B-A3B-Thinking	262K						$0.30	$2
huggingface/XiaomiMiMo/MiMo-V2-Flash	262K						$0.10	$0.30
huggingface/zai-org/GLM-4.7	205K						$0.60	$2
huggingface/zai-org/GLM-4.7-Flash	200K						—	—

Advanced Configuration

Custom Headers

const agent = new Agent({
  name: "custom-agent",
  model: {
    url: "https://router.huggingface.co/v1",
    id: "huggingface/MiniMaxAI/MiniMax-M2.1",
    apiKey: process.env.HF_TOKEN,
    headers: {
      "X-Custom-Header": "value"
    }
  }
});

Dynamic Model Selection

const agent = new Agent({
  name: "dynamic-agent",
  model: ({ runtimeContext }) => {
    const useAdvanced = runtimeContext.task === "complex";
    return useAdvanced
      ? "huggingface/zai-org/GLM-4.7-Flash"
      : "huggingface/MiniMaxAI/MiniMax-M2.1";
  }
});