whatcanirun

maheshbabugorantla/whatcanirun

★ 0 stars Python Code/Dev Tools Updated 1mo ago

MCP server: convert LLM inference budgets into ranked, source-backed plans

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "whatcanirun": {
      "command": "uvx",
      "args": [
        "whatcanirun"
      ]
    }
  }
}

Or install with pip: pip install whatcanirun

README Excerpt

An MCP server that answers the one question every LLM inference user actually asks: > **"I have $X to spend on model Y — what can I actually run?"** Self-hosted, free public APIs only, ships as a stdio MCP server you install with one command. No accounts, no hosting, no telemetry. Given a dollar budget and an open-weight model, returns a ranked, source-backed plan: which GPU on which provider, whether the model fits in VRAM, how many tokens per second you'll get, how many prompts that buys you,

whatcanirun

Quick Install

README Excerpt

Related Code/Dev Tools Servers