Skip to main content

LLM Gateway logoLLM Gateway

Access 181 LLM Gateway models through Mastra's model router. Authentication is handled automatically using the LLMGATEWAY_API_KEY environment variable.

Learn more in the LLM Gateway documentation.

.env
LLMGATEWAY_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: "llmgateway/auto"
});

// 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 LLM Gateway documentation for details.

Models
Direct link to Models

ModelContextToolsReasoningImageAudioVideoInput $/1MOutput $/1M
llmgateway/auto128K——
llmgateway/claude-3-5-sonnet-20241022200K$3$15
llmgateway/claude-3-7-sonnet200K$3$15
llmgateway/claude-3-7-sonnet-20250219200K$3$15
llmgateway/claude-3-opus200K$15$75
llmgateway/claude-haiku-4-5200K$1$5
llmgateway/claude-haiku-4-5-20251001200K$1$5
llmgateway/claude-opus-4-1-20250805200K$15$75
llmgateway/claude-opus-4-20250514200K$15$75
llmgateway/claude-opus-4-5-20251101200K$5$25
llmgateway/claude-opus-4-61.0M$5$25
llmgateway/claude-opus-4-71.0M$5$25
llmgateway/claude-sonnet-4-20250514200K$3$15
llmgateway/claude-sonnet-4-5200K$3$15
llmgateway/claude-sonnet-4-5-20250929200K$3$15
llmgateway/claude-sonnet-4-61.0M$3$15
llmgateway/codestral-2508256K$0.30$0.90
llmgateway/custom128K——
llmgateway/deepseek-r1-052864K$0.80$2
llmgateway/deepseek-v3.1128K$0.56$2
llmgateway/deepseek-v3.2164K$0.28$0.42
llmgateway/devstral-2512262K$0.40$2
llmgateway/devstral-small-2507128K$0.10$0.30
llmgateway/gemini-2.0-flash1.0M$0.10$0.40
llmgateway/gemini-2.0-flash-lite1.0M$0.07$0.30
llmgateway/gemini-2.5-flash1.0M$0.30$3
llmgateway/gemini-2.5-flash-lite1.0M$0.10$0.40
llmgateway/gemini-2.5-flash-lite-preview-09-20251.0M$0.10$0.40
llmgateway/gemini-2.5-pro1.0M$1$10
llmgateway/gemini-3-flash-preview1.0M$0.50$3
llmgateway/gemini-3.1-flash-lite-preview1.0M$0.25$2
llmgateway/gemini-3.1-pro-preview1.0M$2$12
llmgateway/gemini-pro-latest1.0M$2$12
llmgateway/gemma-2-27b-it-together8K$0.08$0.08
llmgateway/gemma-3-12b-it33K——
llmgateway/gemma-3-1b-it1.0M$0.08$0.30
llmgateway/gemma-3-27b128K$0.27$0.27
llmgateway/gemma-3-4b-it33K——
llmgateway/gemma-3n-e2b-it8K——
llmgateway/gemma-3n-e4b-it8K——
llmgateway/glm-4-32b-0414-128k128K$0.10$0.10
llmgateway/glm-4.5131K$0.60$2
llmgateway/glm-4.5-air131K$0.20$1
llmgateway/glm-4.5-airx128K$1$5
llmgateway/glm-4.5-flash131K——
llmgateway/glm-4.5-x128K$2$9
llmgateway/glm-4.5v64K$0.60$2
llmgateway/glm-4.6205K$0.60$2
llmgateway/glm-4.6v128K$0.30$0.90
llmgateway/glm-4.6v-flash128K——
llmgateway/glm-4.6v-flashx128K$0.04$0.40
llmgateway/glm-4.7205K$0.60$2
llmgateway/glm-4.7-flash200K——
llmgateway/glm-4.7-flashx200K$0.07$0.40
llmgateway/glm-5205K$1$3
llmgateway/glm-5.1200K$6$24
llmgateway/gpt-3.5-turbo16K$0.50$2
llmgateway/gpt-48K$30$60
llmgateway/gpt-4-turbo128K$10$30
llmgateway/gpt-4.11.0M$2$8
llmgateway/gpt-4.1-mini1.0M$0.40$2
llmgateway/gpt-4.1-nano1.0M$0.10$0.40
llmgateway/gpt-4o128K$3$10
llmgateway/gpt-4o-mini128K$0.15$0.60
llmgateway/gpt-4o-mini-search-preview128K$0.15$0.60
llmgateway/gpt-4o-search-preview128K$3$10
llmgateway/gpt-5400K$1$10
llmgateway/gpt-5-chat-latest400K$1$10
llmgateway/gpt-5-mini400K$0.25$2
llmgateway/gpt-5-nano400K$0.05$0.40
llmgateway/gpt-5-pro400K$15$120
llmgateway/gpt-5.1400K$1$10
llmgateway/gpt-5.1-codex400K$1$10
llmgateway/gpt-5.1-codex-mini400K$0.25$2
llmgateway/gpt-5.2400K$2$14
llmgateway/gpt-5.2-chat-latest128K$2$14
llmgateway/gpt-5.2-codex400K$2$14
llmgateway/gpt-5.2-pro400K$21$168
llmgateway/gpt-5.3-chat-latest128K$2$14
llmgateway/gpt-5.3-codex400K$2$14
llmgateway/gpt-5.41.1M$3$15
llmgateway/gpt-5.4-mini400K$0.75$5
llmgateway/gpt-5.4-nano400K$0.20$1
llmgateway/gpt-5.4-pro1.1M$30$180
llmgateway/gpt-oss-120b131K$0.15$0.75
llmgateway/gpt-oss-20b131K$0.10$0.50
llmgateway/grok-3131K$3$15
llmgateway/grok-4256K$3$15
llmgateway/grok-4-0709256K$3$15
llmgateway/grok-4-1-fast2.0M$0.20$0.50
llmgateway/grok-4-1-fast-non-reasoning2.0M$0.20$0.50
llmgateway/grok-4-1-fast-reasoning2.0M$0.20$0.50
llmgateway/grok-4-20-beta-0309-non-reasoning2.0M$2$6
llmgateway/grok-4-20-beta-0309-reasoning2.0M$2$6
llmgateway/grok-4-fast2.0M$0.20$0.50
llmgateway/grok-4-fast-non-reasoning2.0M$0.20$0.50
llmgateway/grok-4-fast-reasoning2.0M$0.20$0.50
llmgateway/grok-code-fast-1256K$0.20$2
llmgateway/hermes-2-pro-llama-3-8b8K$0.14$0.14
llmgateway/kimi-k2131K$1$3
llmgateway/kimi-k2-thinking262K$0.60$3
llmgateway/kimi-k2-thinking-turbo262K$1$8
llmgateway/kimi-k2.5262K$0.60$3
llmgateway/llama-3-70b-instruct8K$0.51$0.74
llmgateway/llama-3-8b-instruct8K$0.04$0.04
llmgateway/llama-3.1-70b-instruct128K$0.72$0.72
llmgateway/llama-3.1-8b-instruct128K$0.22$0.22
llmgateway/llama-3.1-nemotron-ultra-253b128K$0.60$2
llmgateway/llama-3.2-11b-instruct128K$0.07$0.33
llmgateway/llama-3.2-3b-instruct33K$0.03$0.05
llmgateway/llama-3.3-70b-instruct128K——
llmgateway/llama-4-maverick-17b-instruct8K$0.24$0.97
llmgateway/llama-4-scout33K$0.18$0.59
llmgateway/llama-4-scout-17b-instruct8K$0.17$0.66
llmgateway/mimo-v2-flash256K$0.10$0.30
llmgateway/minimax-m2197K$0.30$1
llmgateway/minimax-m2.1205K$0.30$1
llmgateway/minimax-m2.1-lightning197K$0.12$0.48
llmgateway/minimax-m2.5205K$0.30$1
llmgateway/minimax-m2.5-highspeed205K$0.60$2
llmgateway/minimax-m2.7205K$0.30$1
llmgateway/minimax-m2.7-highspeed205K$0.60$2
llmgateway/minimax-text-011.0M$0.20$1
llmgateway/ministral-14b-2512262K$0.20$0.20
llmgateway/ministral-3b-2512131K$0.10$0.10
llmgateway/ministral-8b-2512262K$0.15$0.15
llmgateway/mistral-large-2512262K$0.50$2
llmgateway/mistral-large-latest262K$0.50$2
llmgateway/mistral-small-2506128K$0.10$0.30
llmgateway/o1200K$15$60
llmgateway/o3200K$2$8
llmgateway/o3-mini200K$1$4
llmgateway/o4-mini200K$1$4
llmgateway/pixtral-large-latest128K$2$6
llmgateway/qwen-coder-plus131K$0.50$1
llmgateway/qwen-flash1.0M$0.05$0.40
llmgateway/qwen-max33K$2$6
llmgateway/qwen-max-latest33K$2$6
llmgateway/qwen-omni-turbo33K$0.07$0.27
llmgateway/qwen-plus1.0M$0.40$1
llmgateway/qwen-plus-latest131K$0.30$0.90
llmgateway/qwen-turbo1.0M$0.05$0.20
llmgateway/qwen-vl-max131K$0.80$3
llmgateway/qwen-vl-plus131K$0.21$0.63
llmgateway/qwen2-5-vl-32b-instruct131K$0.30$0.30
llmgateway/qwen2-5-vl-72b-instruct131K$3$8
llmgateway/qwen25-coder-7b131K$0.05$0.05
llmgateway/qwen3-235b-a22b-fp8131K$0.50$3
llmgateway/qwen3-235b-a22b-instruct-2507131K$0.80$2
llmgateway/qwen3-235b-a22b-thinking-2507131K$0.80$2
llmgateway/qwen3-30b-a3b-fp8131K$0.10$0.10
llmgateway/qwen3-30b-a3b-instruct-2507131K$0.10$0.10
llmgateway/qwen3-30b-a3b-thinking-2507131K$0.10$0.10
llmgateway/qwen3-32b131K$0.70$3
llmgateway/qwen3-32b-fp8131K$0.10$0.10
llmgateway/qwen3-4b-fp8131K$0.03$0.05
llmgateway/qwen3-coder-30b-a3b-instruct262K$0.45$2
llmgateway/qwen3-coder-480b-a35b-instruct262K$2$8
llmgateway/qwen3-coder-flash1.0M$0.30$2
llmgateway/qwen3-coder-next262K$0.80$4
llmgateway/qwen3-coder-plus1.0M$1$5
llmgateway/qwen3-max262K$1$6
llmgateway/qwen3-max-2026-01-23256K$3$15
llmgateway/qwen3-next-80b-a3b-instruct131K$0.50$2
llmgateway/qwen3-next-80b-a3b-thinking131K$0.50$6
llmgateway/qwen3-vl-235b-a22b-instruct131K$0.80$2
llmgateway/qwen3-vl-235b-a22b-thinking131K$0.80$2
llmgateway/qwen3-vl-30b-a3b-instruct131K$0.10$0.10
llmgateway/qwen3-vl-30b-a3b-thinking131K$0.10$0.10
llmgateway/qwen3-vl-8b-instruct131K$0.10$0.10
llmgateway/qwen3-vl-flash1.0M$0.05$0.40
llmgateway/qwen3-vl-plus262K$0.20$2
llmgateway/qwen35-397b-a17b262K$0.60$4
llmgateway/qwq-plus131K$0.80$2
llmgateway/seed-1-6-250615256K$0.25$2
llmgateway/seed-1-6-250915256K$0.25$2
llmgateway/seed-1-6-flash-250715256K$0.07$0.30
llmgateway/seed-1-8-251228256K$0.25$2
llmgateway/sonar128K$1$1
llmgateway/sonar-pro200K$3$15
llmgateway/sonar-reasoning-pro128K$2$8
181 available models

Advanced configuration
Direct link to Advanced configuration

Custom headers
Direct link to Custom headers

src/mastra/agents/my-agent.ts
const agent = new Agent({
  id: "custom-agent",
  name: "custom-agent",
  model: {
    url: "https://api.llmgateway.io/v1",
    id: "llmgateway/auto",
    apiKey: process.env.LLMGATEWAY_API_KEY,
    headers: {
      "X-Custom-Header": "value"
    }
  }
});

Dynamic model selection
Direct link to 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
      ? "llmgateway/sonar-reasoning-pro"
      : "llmgateway/auto";
  }
});
On this page