Skip to content

MCP servers and agent skills for gpu

38 MCP servers and 13 agent skills for gpu in AI models and LLM tooling, including Agent platform deploy, ML engineering and Huggingface zerogpu, most popular first. 35 of the 51 are listed as free or open source. Each one is written up from its own site or README, with pricing, platforms and alternatives.

Showing 31–51 of 51

  1. Estimate model weight memory, facility energy and token cost with read-only calculators

    Not checked yet
    • 3 tools
    • Remote
    • Free
    • AI models and LLM tooling
  2. Local AI Picks

    MCP server

    Find open-weight AI models that fit your memory, with file sizes, sources and test status

    Not checked yet
    • 5 tools
    • Remote
    • Free
    • AI models and LLM tooling
  3. Compare GPUs, Macs and mini PCs by which open models they run and what they cost today

    Not checked yet
    • Remote
    • Free
    • AI models and LLM tooling
  4. Auxen

    MCP server

    Hire scheduled AI workers or provision private model instances over MCP

    Not checked yet
    • 7 tools
    • Remote
    • Freemium
    • AI models and LLM tooling
  5. Directional answers on whether quantizing an LLM saves GPU energy, with stated basis

    Not checked yet
    • Remote
    • Free
    • AI models and LLM tooling
  6. Proxlai

    MCP server

    Dated, sourced prices for AI APIs, GPUs and more, searchable by agents

    Not checked yet
    • 5 tools
    • Remote
    • Freemium
    • Design and media
  7. VRAMGlass

    MCP server

    Free MCP server for live GPU prices, price history and local LLM VRAM requirements

    Not checked yet
    • Remote
    • Freemium
    • AI models and LLM tooling
  8. aicharged.io

    MCP server

    One MCP endpoint with 317 tools for images, video, audio, 3D, data, storage, mail and GPUs

    Not checked yet
    • Remote
    • Freemium
    • Design and media
  9. Ask which GPU a model needs, whether it fits, and what renting it costs

    Not checked yet
    • 5 tools
    • Remote
    • Free
    • AI models and LLM tooling
  10. Lead Green

    MCP server

    GPU energy per million output tokens by model, with carbon estimates by country

    Runs locally
    • 8 tools
    • Open source
    • AI models and LLM tooling
  11. Guaardvark

    MCP server

    Local MCP server giving editors image, video, RAG and code-map tools from Guaardvark

    Runs locally
    • 28 tools
    • Open source
    • Design and media
  12. Neurarch

    MCP server

    Size GPU training jobs and lint PyTorch models inside your coding agent, offline

    Runs locally
    • 7 tools
    • Free
    • AI models and LLM tooling
  13. Model Tested

    MCP server

    Independent open-weight LLM test scores: which local model fits your GPU or Mac

    Not checked yet
    • 5 tools
    • Remote
    • Free
    • AI models and LLM tooling
  14. Computegpu

    MCP server

    Rent cloud GPUs from your agent: browse the catalogue, deploy, stop, destroy

    Not checked yet
    • Remote
    • Paid
    • Cloud and infrastructure
  15. supercomputer

    MCP server

    Run GPU jobs on L4, A100 or H100 machines from a chat, billed per second

    Not checked yet
    • 9 tools
    • Remote
    • Paid
    • Cloud and infrastructure
  16. Dated catalogue of free AI API tiers, trial credits and discounts, plus API health probes

    Not checked yet
    • 3 tools
    • Remote
    • Free
    • AI models and LLM tooling
  17. Check whether an LLM fits your GPU, with VRAM, KV cache and GPU counts

    Not checked yet
    • Remote
    • Free
    • AI models and LLM tooling
  18. GPU Cloud

    MCP server

    Rent NVIDIA GPU servers by the hour in Montréal, billed in Canadian dollars

    Not checked yet
    • Remote
    • Paid
    • Cloud and infrastructure
  19. Symbioza

    MCP server

    Run containerized GPU jobs under a hard dollar cap and collect the output files

    Not checked yet
    • 9 tools
    • Remote
    • Paid
    • Cloud and infrastructure
  20. Check whether a GPU or Mac can run a local LLM: memory fit, speed estimate, model specs

    Not checked yet
    • Remote
    • Free
    • AI models and LLM tooling
  21. MCP

    MCP server

    Rent NVIDIA GPUs from an agent and pay per second onchain

    Runs locally
    • 20 tools
    • Paid
    • Cloud and infrastructure