Gpu Bridge Mcp Server

Name: Gpu Bridge Mcp Server
Author: gpu-bridge

@gpu-bridge

关于 Gpu Bridge Mcp Server

Access 47 AI models and 30 services via MCP. LLM, image gen, video, speech, embeddings, reranking, PDF parsing & more. Pay-per-use with x402 (USDC) or API key.

#ai #mcp #gpu #llm #inference

基本信息

分类

开发工具

传输方式

stdio

发布者

gpu-bridge

提交者

fjnunezp75

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "gpu-bridge": {
      "command": "npx",
      "args": [
        "-y",
        "@gpu-bridge/mcp-server"
      ],
      "env": {
        "GPUBRIDGE_API_KEY": "your-api-key"
      }
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Gpu Bridge Mcp Server?

Gpu Bridge Mcp Server is an MCP server that exposes 30 GPU-powered AI services (LLMs, image generation, audio, video, embeddings, reranking, PDF parsing, NSFW detection, and more) as Model Context Protocol tools. It integrates with the GPU-Bridge API and supports the x402 payment protocol, allowing autonomous AI agents to pay per request on-chain with USDC on Base L2 without API keys or accounts.

How to use Gpu Bridge Mcp Server?

Install by adding a configuration entry to claude_desktop_config.json using npx -y @gpu-bridge/mcp-server and setting the GPUBRIDGE_API_KEY environment variable (or use x402 for keyless agent payments). Once configured, Claude Desktop can invoke services through the MCP tools: gpu_run, gpu_catalog, gpu_estimate, gpu_status, and gpu_balance.

Key features of Gpu Bridge Mcp Server

30 GPU-powered AI services as MCP tools
Native x402 support for autonomous agent payments
No API keys or accounts required with x402
Pay per request using USDC on Base L2
Tools for executing, cataloging, estimating, and checking job status
Services include LLMs, image gen, audio, video, embeddings, and more

Use cases of Gpu Bridge Mcp Server

AI agents autonomously paying for LLM inference without human intervention
Generating images from text prompts (e.g., FLUX, Stable Diffusion)
Transcribing audio with speaker diarization
Running code generation or code embeddings
Parsing PDFs and extracting text from documents

FAQ from Gpu Bridge Mcp Server

How do I get an API key for Gpu Bridge Mcp Server?

Visit gpubridge.xyz and grab a free API key, or use the x402 protocol for keyless agent payments.

What is the x402 protocol and how does it work?

x402 is an open payment protocol that enables autonomous AI agents to pay for compute with USDC on Base L2. The agent receives an HTTP 402 Payment Required response, pays USDC (gas < $0.01, settles in 2 seconds), then retries with a payment proof to get the result.

What MCP tools does Gpu Bridge Mcp Server provide?

It provides five tools: gpu_run (execute any service), gpu_catalog (list available services with pricing), gpu_estimate (estimate cost before running), gpu_status (check job status and results), and gpu_balance (check balance and spend).

How much do the services cost?

Pricing starts from $0.003/1K tokens for LLMs, $0.01/image for image generation, $0.005/minute for speech-to-text, $0.005/1K chars for TTS, $0.0001/1K tokens for embeddings, $0.001/query for reranking, and $0.005/document for PDF parsing. All prices are in USD; x402 payments are in USDC on Base L2.

Do I need an account to use Gpu Bridge Mcp Server?

No, with the x402 protocol you can pay per request on-chain without any account or API key. Alternatively, you can use a free API key from gpubridge.xyz.