NanoGPT

Access 21 NanoGPT models through Mastra's model router. Authentication is handled automatically using the NANO_GPT_API_KEY environment variable.

Learn more in the NanoGPT documentation.

NANO_GPT_API_KEY=your-api-key

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

const agent = new Agent({
  name: "my-agent",
  instructions: "You are a helpful assistant",
  model: "nano-gpt/deepseek/deepseek-r1"
});

// 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 NanoGPT documentation for details.

Models

21 available models

Model	Context	Tools	Reasoning	Image	Audio	Video	Input $/1M	Output $/1M
nano-gpt/deepseek/deepseek-r1	128K						$1	$2
nano-gpt/deepseek/deepseek-v3.2:thinking	128K						$1	$2
nano-gpt/meta-llama/llama-3.3-70b-instruct	128K						$1	$2
nano-gpt/meta-llama/llama-4-maverick	128K						$1	$2
nano-gpt/minimax/minimax-m2.1	128K						$1	$2
nano-gpt/mistralai/devstral-2-123b-instruct-2512	131K						$1	$2
nano-gpt/mistralai/ministral-14b-instruct-2512	131K						$1	$2
nano-gpt/mistralai/mistral-large-3-675b-instruct-2512	131K						$1	$2
nano-gpt/moonshotai/kimi-k2-instruct	131K						$1	$2
nano-gpt/moonshotai/kimi-k2-thinking	33K						$1	$2
nano-gpt/nousresearch/hermes-4-405b:thinking	128K						$1	$2
nano-gpt/nvidia/llama-3_3-nemotron-super-49b-v1_5	128K						$1	$2
nano-gpt/openai/gpt-oss-120b	128K						$1	$2
nano-gpt/qwen/qwen3-235b-a22b-thinking-2507	262K						$1	$2
nano-gpt/qwen/qwen3-coder	106K						$1	$2
nano-gpt/z-ai/glm-4.6	200K						$1	$2
nano-gpt/z-ai/glm-4.6:thinking	128K						$1	$2
nano-gpt/zai-org/glm-4.5-air	128K						$1	$2
nano-gpt/zai-org/glm-4.5-air:thinking	128K						$1	$2
nano-gpt/zai-org/glm-4.7	205K						$1	$2
nano-gpt/zai-org/glm-4.7:thinking	128K						$1	$2

Advanced Configuration

Custom Headers

const agent = new Agent({
  name: "custom-agent",
  model: {
    url: "https://nano-gpt.com/api/v1",
    id: "nano-gpt/deepseek/deepseek-r1",
    apiKey: process.env.NANO_GPT_API_KEY,
    headers: {
      "X-Custom-Header": "value"
    }
  }
});

Dynamic Model Selection

const agent = new Agent({
  name: "dynamic-agent",
  model: ({ runtimeContext }) => {
    const useAdvanced = runtimeContext.task === "complex";
    return useAdvanced
      ? "nano-gpt/zai-org/glm-4.7:thinking"
      : "nano-gpt/deepseek/deepseek-r1";
  }
});