Neuralwatt

Access 31 Neuralwatt models through Mastra's model router. Authentication is handled automatically using the NEURALWATT_API_KEY environment variable.

Learn more in the Neuralwatt documentation.

.env

NEURALWATT_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: "neuralwatt/deepseek-v4-flash"
});

// 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 Neuralwatt documentation for details.

Models

23 available models

Model	Context	Tools	Reasoning	Image	Audio	Video	Input $/1M	Output $/1M
neuralwatt/deepseek-v4-flash	1.0M						$0.14	$0.28
neuralwatt/deepseek-v4-flash-flex	1.0M						$0.09	$0.18
neuralwatt/deepseek-v4-flash-speed	1.0M						$0.14	$0.28
neuralwatt/deepseek-v4.1-flash	1.0M						$0.15	$0.60
neuralwatt/deepseek-v4.1-flash-flex	1.0M						$0.10	$0.39
neuralwatt/gemma-4-31b	262K						$0.14	$0.42
neuralwatt/glm-5.3	1.0M						$1	$5
neuralwatt/glm-5.3-flash	1.0M						$0.15	$0.50
neuralwatt/glm-5.3-flash-flex	1.0M						$0.10	$0.33
neuralwatt/glm-5.3-flex	1.0M						$0.94	$3
neuralwatt/kimi-k2.7-code	262K						$0.95	$4
neuralwatt/kimi-k2.7-code-fast	262K						$0.95	$4
neuralwatt/kimi-k2.7-code-flex	262K						$0.62	$3
neuralwatt/kimi-k3	1.0M						$3	$15
neuralwatt/kimi-k3-fast	1.0M						$3	$15
neuralwatt/kimi-k3-flex	1.0M						$2	$10
neuralwatt/mimo-v2.6-pro	1.0M						$0.87	$2
neuralwatt/mimo-v2.6-pro-flex	1.0M						$0.57	$1
neuralwatt/qwen-3.8-27b	262K						$0.45	$3
neuralwatt/qwen-3.8-27b-flex	262K						$0.29	$2
neuralwatt/qwen3.6-35b	131K						$0.29	$1
neuralwatt/qwen3.6-35b-fast	131K						$0.29	$1
neuralwatt/qwen3.6-35b-flex	131K						$0.19	$0.75

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

Advanced configuration

Custom headers

src/mastra/agents/my-agent.ts

const agent = new Agent({
  id: "custom-agent",
  name: "custom-agent",
  model: {
    url: "https://api.neuralwatt.com/v1",
    id: "neuralwatt/deepseek-v4-flash",
    apiKey: process.env.NEURALWATT_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
      ? "neuralwatt/qwen3.6-35b-flex"
      : "neuralwatt/deepseek-v4-flash";
  }
});