Gpu Bridge

io.github.fjnunezp75由 GitHub 账号 fjnunezp75 发布v2.4.3更新于 Oct 24, 2026

30 GPU-powered AI services as MCP tools. LLM, image, video, audio, embeddings & more.

已验证STDIO仅桌面媒体与设计AI 与机器学习

概览

AI 生成的概览

把 30 项付费 GPU 推理服务(大模型、图像、视频、音频、向量与 PDF 解析)封装为 MCP 工具,供助手调用并查看结果。

可以做什么
  • 按服务键运行任意 GPU-Bridge 服务,可指定输入参数与 fast 或 cheap 优先级
  • 查看服务目录及其价格与能力,并在运行前估算费用
  • 查询任务状态并获取结果
  • 查看余额、每日支出与用量折扣档位
  • 覆盖大语言模型、图像生成与分析、语音转写、语音合成、音乐与音效、向量嵌入、重排序、视频与 PDF 解析
适用场景
适合希望助手直接调用托管 GPU 模型完成文本、图像、音频、视频、向量或文档解析任务,而不自行维护推理基础设施的场景,尤其是按次计费可接受的零散任务。
运行要求
通过 npx 从 npm 包 @gpu-bridge/mcp-server 在本地运行,需写入桌面客户端配置文件。需要 GPUBRIDGE_API_KEY 环境变量中的 GPU-Bridge API 密钥,或改用 x402 微支付。需要访问 GPU-Bridge API 的网络连接。
安装前请注意
每次运行都按次以美元计费,使用越多花费越高,建议先估算并查看余额。API 密钥是需在环境变量中传递的机密。x402 方式需用钱包私钥在链上以 USDC 付款,私钥属高价值机密,务必妥善保护。提示词、媒体与文档会发送给第三方服务。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Gpu Bridge,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

GPU-Bridge MCP Server

30 GPU-powered AI services as MCP tools — LLMs, image generation, audio, video, embeddings, reranking, PDF parsing, NSFW detection & more. x402 native for autonomous AI agents: pay per request on-chain with USDC on Base L2. No API keys. No accounts.

[npm version] [License: MIT] [MCP Compatible]

What is GPU-Bridge?

GPU-Bridge is a unified GPU inference API with native x402 support — the open payment protocol that allows AI agents to autonomously pay for compute with USDC on Base L2. No API keys, no accounts, no human intervention required.

This MCP server exposes all 30 GPU-Bridge services as Model Context Protocol tools, giving Claude (and any MCP-compatible AI) direct access to GPU inference.


Install in Claude Desktop (2 minutes)

1. Get your API key (or use x402 for autonomous agents)

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

2. Add to claude_desktop_config.json

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

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

3. Restart Claude Desktop

That's it. Claude now has access to 30 GPU-powered AI services.


MCP Tools

gpu_run

Run any GPU-Bridge service. The primary tool for executing AI tasks.

Parameters:  service  (string)  — Service key (e.g., "llm-4090", "flux-schnell", "whisper-l4")  input    (object)  — Service-specific input parameters  priority (string)  — Optional: "fast" (lowest latency) or "cheap" (lowest cost)

gpu_catalog

Get the full catalog of available services with pricing and capabilities.

gpu_estimate

Estimate cost before running a service. No authentication required.

gpu_status

Check the status of a job and retrieve results.

gpu_balance

Check your current balance, daily spend, and volume discount tier.


30 Available Services

Language Models (LLMs)

Service IDDescriptionNotes
llm-4090General purpose LLMSub-second via Groq
llm-a100Maximum capability LLMLargest models
llm-l4Ultra-fast, low cost LLMBudget option
code-4090Code generationOptimized for code
llm-streamStreaming LLM responsesReal-time output

Image Generation

Service IDDescriptionNotes
flux-schnellFLUX.1 SchnellFast, 4-step generation
flux-devFLUX.1 DevHigh quality
sdxl-4090Stable Diffusion XLVersatile
sd35-l4Stable Diffusion 3.5Latest SD model
img2img-4090Image-to-imageStyle transfer, editing

Vision & Image Analysis

Service IDDescriptionNotes
llava-4090Visual Q&AImage understanding
ocr-l4Text extraction (OCR)Multi-language
rembg-l4Background removalInstant
caption-4090Image captioningAuto-describe images
nsfw-detectContent moderationNSFW classification

Speech-to-Text

Service IDDescriptionNotes
whisper-l4Fast transcriptionSub-second
whisper-a100High accuracy transcriptionLarge files
diarize-l4Speaker diarizationWho said what

Text-to-Speech

Service IDDescriptionNotes
tts-l4Voice cloning TTS40+ voices
tts-fastUltra-fast TTSLowest latency
bark-4090Expressive TTSEmotion, laughter

Audio Generation

Service IDDescriptionNotes
musicgen-l4Music generationText-to-music
audiogen-l4Sound effectsText-to-SFX

Embeddings & Search

Service IDDescriptionNotes
embed-l4Text embeddingsMultilingual
embed-codeCode embeddingsFor code search
rerankDocument rerankingJina, sub-second

Video

Service IDDescriptionNotes
animatediffText-to-videoAnimateDiff
video-enhanceVideo upscalingUp to 4K

Utilities

Service IDDescriptionNotes
pdf-parseDocument parsingPDF/DOCX to text

x402: For Autonomous AI Agents

GPU-Bridge supports the x402 payment protocol, enabling truly autonomous AI agents to pay for compute without human intervention.

Agent Request → GPU-Bridge returns HTTP 402 Payment Required      ↓Agent pays USDC on Base L2 (gas < $0.01, settles in 2s)      ↓Agent retries with payment proof → GPU-Bridge executes and returns result

Python Example with x402

python
from x402.client import PaymentClient
client = PaymentClient(private_key="0x...", chain="base")
response = client.request(    "POST",    "https://api.gpubridge.io/v1/run",    json={        "service": "flux-schnell",        "input": {"prompt": "A robot painting on a canvas", "steps": 4}    })print(response.json())

Pricing

CategoryStarting From
LLMs$0.003/1K tokens
Image Generation$0.01/image
Speech-to-Text$0.005/minute
Text-to-Speech$0.005/1K chars
Embeddings$0.0001/1K tokens
Reranking$0.001/query
PDF Parsing$0.005/document

All prices in USD. x402 payments in USDC on Base L2.


Links


License

MIT © Healthtech Capital LLC

来源:README.md,提交 033b2c1

工具

0
SourceWeft 尚未在沙箱中启动过该软件包,因此其工具未知。

版本历史

1
  1. v2.4.3最新Sep 16, 2026

评分

已满足适用检查项的 50%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分

    请保持端点能够响应 MCP 握手;SourceWeft 每 30 天探测一次。

  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头,或保持端点能够响应探测,以便判定其认证方式。

  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分