# mcp-server

Access **47 AI models** and **30 services** via a single MCP tool.

## Quick Start

```bash
# Connect this server (installs CLI if needed)
npx -y @smithery/cli@latest mcp add gpu-bridge/mcp-server

# Browse available tools
npx -y @smithery/cli@latest tool list gpu-bridge/mcp-server

# Get full schema for a tool
npx -y @smithery/cli@latest tool get gpu-bridge/mcp-server gpu_run

# Call a tool
npx -y @smithery/cli@latest tool call gpu-bridge/mcp-server gpu_run '{}'
```

## Direct MCP Connection

Endpoint: `https://mcp-server--gpu-bridge.run.tools`

**Optional config:**
- `apiKey` (header) — GPU-Bridge API key (gpub_...). Free at gpubridge.xyz. Needed for gpu_run.

## Tools (5)

- `gpu_run` — Run any GPU-Bridge AI service. 30 services available: LLM inference (sub-second), image generation (FLUX, SD3.5), video…
- `gpu_catalog` — List all available GPU-Bridge services with pricing and model info.
- `gpu_status` — Check the status of a GPU-Bridge job and retrieve results.
- `gpu_balance` — Check GPU-Bridge credit balance, daily spend, and volume discount tier.
- `gpu_estimate` — Estimate the cost of a GPU-Bridge service before running it.

```bash
# Get full input/output schema for a tool
npx -y @smithery/cli@latest tool get gpu-bridge/mcp-server <tool-name>
```