How to use Mesh LLM: what you can do with it

3 use cases

• Run a local model behind an OpenAI-compatible API in one command

  Install Mesh LLM, let it pick a model and backend, and point any OpenAI client at localhost.

  • REPO Mesh-LLM/mesh-llm
  • Software and DevOps
  • Beginner
  View prompt

  Install Mesh-LLM/mesh-llm on this machine, run setup and serve --auto, list the available models, and send one test chat completion to localhost:9337 so I can point <app> at it.

  Share Email me this

• Pool two home machines into a private model mesh

  Start a private mesh on one machine and join the others with an invite token so they share one API.

  • REPO Mesh-LLM/mesh-llm
  • Software and DevOps
  • Intermediate
  View prompt

  Set up a private Mesh-LLM/mesh-llm mesh across <machine A> and <machine B>: start it with a Qwen3-8B model on the first, join the second by invite token, and confirm both appear in the console.

  Share Email me this

• Run Goose, OpenCode, Claude Code or Pi against your own mesh

  Use Mesh LLM's agent launchers so coding agents run on models served by your own machines.

  • REPO Mesh-LLM/mesh-llm
  • Claude Code
  • Software and DevOps
  • Intermediate
  View prompt

  Start Mesh-LLM/mesh-llm and launch <OpenCode or Goose or Claude Code or Pi> through it, choose the best model the mesh serves for coding, and run a small test task in <repo>.

  Share Email me this

What people are building with Mesh LLM

• Hands-onthe web

  Linked a CPU server, a rented 5090 pod and a Mac into one Mesh LLM endpoint

  The team made the always on CPU box the hub serving Qwen3 0.6B and joined the Mac as a client only node with an invite token. A curl from the Mac to localhost:9337/v1 was answered by the box. The post lists the snags they hit, including an internal docker address in the invite token.

  Mesh LLM on GitHub

  By Yim (productize.life) on the web·Jul 25, 2026

  View original post on the web

• Hands-onHacker News

  Joined the public Mesh LLM swarm from a MacBook Pro with one command and it worked first try

  A Hacker News commenter ran mesh-llm --auto from the swarm page join button. It picked a model to serve, downloaded it from peers, and he chatted with the model he was hosting while watching his GPU do the work.

  Mesh LLM on GitHubTry it: Run Goose, OpenCode, Claude Code or Pi against your own mesh

  By maccam912 on Hacker News·Jul 12, 2026

  View original post on Hacker News

Real builds shared on X, Hacker News, Reddit, DEV and GitHub, summed up by Deduped from each post and its links; each card links to the original post. See what people are building with other repos

Quick start with Mesh LLM

1. Check you have what it needs: macOS, Linux or Windows (install script, PowerShell or Homebrew on Apple Silicon); One or more machines with a GPU or enough memory for the model; curl and jq for the README's test requests.
2. Open Mesh-LLM/mesh-llm on GitHub and read the README and license.
3. Install with curl -fsSL https://raw.githubusercontent.com/Mesh-LLM/mesh-llm/main/install.sh | bash
4. Finish setup with mesh-llm setup
5. Start serving with mesh-llm serve --auto
6. List models with curl -s http://localhost:9337/v1/models | jq '.data[].id'

These steps come from our top use case, Run a local model behind an OpenAI-compatible API in one command, which also has a copy-paste agent prompt.

Using the Deduped MCP, get the setup steps for Mesh-LLM/mesh-llm

claude mcp add --transport http deduped https://dedupedai.com/mcp

Best AI repos for Software and DevOps

About this repo

• Who it is forHome lab owners and small teams with several computers who want to run models bigger than any one of them can hold.
• Why it is hot3.6k stars, talked about in 3 AI sources we track
• LanguageRust
• Sourcegithub.com/Mesh-LLM/mesh-llm · Back to Hot repos

Requirements

• macOS, Linux or Windows (install script, PowerShell or Homebrew on Apple Silicon)
• One or more machines with a GPU or enough memory for the model
• curl and jq for the README's test requests

Related repos

• Trended earlier REPO L-Forster/open-jet Open Jet OpenJet is a local AI coding agent and LLM inference engine that runs on your own GPU with no account or API key. `openjet setup` measures your hardware, picks and downloads a model and builds llama.cpp, and on RTX 30, 40 and 50 cards compiles its own CUDA megakernel for Qwen3.8-27B. It can serve 4 agents on one GPU, pair a frontier planner with a local worker (Slipstream), use Ollama or LM Studio, and ship a local model in your app via a Python SDK.
• Trended earlier REPO simonw/llm LLM LLM is a CLI tool and Python library for interacting with OpenAI, Gemini, Qwen, DeepSeek, Mistral and dozens of other large language models, through remote APIs or models installed on your own machine. It can run prompts from the command line, store prompts and responses in SQLite, generate embeddings, extract structured content, and let models execute tools.
• Trended earlier REPO mudler/LocalAI LocalAI LocalAI is an open-source AI engine that runs LLMs, vision, voice, image and video models on any hardware, with no GPU required. Each backend (llama.cpp, vLLM, whisper.cpp, stable-diffusion, MLX and more) is pulled only when a model needs it, and every backend sits behind drop-in OpenAI, Anthropic and ElevenLabs compatible APIs, with multi-user auth, quotas and built-in agents with tools, RAG and MCP.