TokenGo

Access 13 TokenGo models through Mastra's model router. Authentication is handled automatically using the TOKENGO_API_KEY environment variable.

Learn more in the TokenGo documentation.

.env

TOKENGO_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: "tokengo/deepseek/deepseek-v3.1"
});

// 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);
}

note

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

Models

13 available models

Model	Context	Tools	Reasoning	Image	Audio	Video	Input $/1M	Output $/1M
tokengo/deepseek/deepseek-v3.1	131K						$0.19	$0.71
tokengo/deepseek/deepseek-v3.2	128K						$0.22	$0.33
tokengo/deepseek/deepseek-v4-flash	1.0M						$0.10	$0.20
tokengo/deepseek/deepseek-v4-pro	1.0M						$0.43	$0.87
tokengo/minimax/minimax-m2.5	205K						$0.30	$1
tokengo/moonshotai/kimi-k2.6	262K						$0.95	$4
tokengo/moonshotai/kimi-k3	1.0M						$3	$15
tokengo/qwen/qwen3.5-397b-a17b	262K						$0.40	$3
tokengo/z-ai/glm-5	205K						$0.89	$3
tokengo/z-ai/glm-5.1	200K						$1	$4
tokengo/z-ai/glm-5.2	1.0M						$1	$4
tokengo/z-ai/glm-5.3	1.0M						$1	$4
tokengo/z-ai/glm-5.3-flash	1.0M						$0.07	$0.03

Advanced configuration

Custom headers

src/mastra/agents/my-agent.ts

const agent = new Agent({
  id: "custom-agent",
  name: "custom-agent",
  model: {
    url: "https://api.tokengo.com/v1",
    id: "tokengo/deepseek/deepseek-v3.1",
    apiKey: process.env.TOKENGO_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
      ? "tokengo/z-ai/glm-5.3-flash"
      : "tokengo/deepseek/deepseek-v3.1";
  }
});