广告合作
MCP.so
登录

FitLLM MCP Server

@click6067-ship-it

代码仓库主页文档

基本信息

分类

开发工具

许可证

MIT

作者

click6067-ship-it

提交者

Yo

关于 FitLLM MCP Server

Will this LLM fit on your GPU, multi-GPU rig or Mac? Exact VRAM & KV-cache math. Remote server: https://fitllm.run/api/mcp

#llm#vram#gpu#hardware#apple-silicon#quantization#multi-gpu#calculator

FitLLM MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "fitllm": {
      "url": "https://fitllm.run/api/mcp"
    }
  }
}

FitLLM MCP Server 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表，不代表服务没有工具。请查阅项目文档了解支持的能力。

点击上方按钮,直接连接服务器端点实时获取工具列表。

概览

FitLLM MCP

Read-only, architecture-aware memory estimates for local LLMs on GPUs, multi-GPU rigs, and Apple Silicon Macs. The public server is stateless and requires no login.

Remote MCP

{"mcpServers":{"fitllm":{"url":"https://fitllm.run/api/mcp"}}}

Tools

  • check_llm_fit checks one model and returns a fit, tight, or no-fit verdict with a memory breakdown.
  • what_fits_on_hardware ranks supported models for a specified GPU, rig, or Mac.
  • list_supported lists accepted built-in model and hardware names.

Evidence and limits

Results are estimates computed by the open MIT-licensed engine from modeled architecture fields and documented hardware data. Actual use varies by runtime, quantization, driver, and OS state. Unsupported architectures fail closed instead of publishing a guess.

Source and reproducible tests: https://github.com/click6067-ship-it/fitllm-engine Interactive calculator: https://fitllm.run

评论

开发工具 分类下的更多 MCP 服务器

DocBin

DocBin

Give an agent a place to put large output. Instead of returning 40k characters of a build log into the conversation, it posts the file to DocBin and hands back one short URL. Docs render with syntax highlighting and mark

Foremerge

naw103

Catch intent conflicts before code conflicts. The open-source coordination protocol for coding agents, built above Git.

50

OrangePro

Aamir Siddiqui

OrangePro local-first CLI + MCP server for behavior mapping, grounded test generation, and dynamic proof.

12

Raccha AI

Shree Mandadi

MCP-first toolbox for agents: namespaced key-value storage, a FIFO queue, auth, and a handful of stateless utility tools (JWT decode, hashing, cert inspection, CIDR math). Free while in early access. This repository is a

FARPY

mangomunchr

Verified distributed compute for humans, applications, and agents. Submit Blender, Octane, and video workloads, route them across qualified GPUs, track asynchronous execution, download finished artifacts, and verify FARP

1
广告合作