# AI use cases for Mesh LLM (Mesh-LLM/mesh-llm) Repo: [Mesh-LLM/mesh-llm](https://github.com/Mesh-LLM/mesh-llm) | Use cases: 3 | Page: https://dedupedai.com/use-cases/mesh-llm-mesh-llm/ Mesh LLM pools GPUs and memory across machines and exposes them as one OpenAI-compatible API at localhost:9337/v1. You start one node, add more later, and the mesh decides whether a model runs locally, routes to a peer, or is split across machines in layer stages. Traffic between nodes is end-to-end encrypted over QUIC. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Run a local model behind an OpenAI-compatible API in one command: https://dedupedai.com/use-cases/mesh-llm-mesh-llm/serve-a-local-model-through-an-openai-compatible-api/ - Pool two home machines into a private model mesh: https://dedupedai.com/use-cases/mesh-llm-mesh-llm/pool-two-home-machines-into-a-private-mesh/ - Run Goose, OpenCode, Claude Code or Pi against your own mesh: https://dedupedai.com/use-cases/mesh-llm-mesh-llm/point-claude-code-or-opencode-at-your-mesh/ ## How to run a local model behind an OpenAI-compatible API in one command with Mesh LLM Repo: [Mesh-LLM/mesh-llm](https://github.com/Mesh-LLM/mesh-llm) | Difficulty: Beginner | Time: 20 minutes | Basis: From the README Install Mesh LLM, let it pick a model and backend, and point any OpenAI client at localhost. **Who it is for:** A developer who wants a local endpoint for apps that already speak the OpenAI API. **The problem:** Setting up a backend, downloading a model and exposing an API by hand takes several tools. **The outcome:** A local /v1 API on port 9337 and a web console on port 3131, ready for chat completions. ### Steps 1. Install with `curl -fsSL https://raw.githubusercontent.com/Mesh-LLM/mesh-llm/main/install.sh | bash` 2. Finish setup with `mesh-llm setup` 3. Start serving with `mesh-llm serve --auto` 4. List models with `curl -s http://localhost:9337/v1/models | jq '.data[].id'` 5. Send a chat completion request to http://localhost:9337/v1/chat/completions from your app ### Prompt to paste into your agent ```text 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 at it. ``` ### Caveat This card is written from the Mesh-LLM/mesh-llm README, not from a test run by Deduped. Before you start, make sure you have macOS, Linux or Windows. Check the repo for newer instructions before you rely on these steps. Risk, remote script: The quick start pipes install.sh into bash; use the Homebrew formula on Apple Silicon if you prefer. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/mesh-llm-mesh-llm/serve-a-local-model-through-an-openai-compatible-api/ ## How to pool two home machines into a private model mesh with Mesh LLM Repo: [Mesh-LLM/mesh-llm](https://github.com/Mesh-LLM/mesh-llm) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Start a private mesh on one machine and join the others with an invite token so they share one API. **Who it is for:** A home lab owner with a gaming PC and a Mac who wants both to serve models together. **The problem:** Each machine alone is too small for the model you want, and they cannot share work by default. **The outcome:** One private mesh where every node exposes the same /v1 API and requests route to the node that can serve them. ### Steps 1. On the first machine run `mesh-llm serve --model Qwen3-8B-Q4_K_M` 2. Get the invite token and run the join command on each other machine 3. Open the web console to see which node serves which model 4. For a model too big for one box, use the split serving guide to run layer stages across machines ### Prompt to paste into your agent ```text Set up a private Mesh-LLM/mesh-llm mesh across and : start it with a Qwen3-8B model on the first, join the second by invite token, and confirm both appear in the console. ``` ### Caveat This card is written from the Mesh-LLM/mesh-llm README, not from a test run by Deduped. Before you start, make sure you have macOS, Linux or Windows. Check the repo for newer instructions before you rely on these steps. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/mesh-llm-mesh-llm/pool-two-home-machines-into-a-private-mesh/ ## How to run Goose, OpenCode, Claude Code or Pi against your own mesh with Mesh LLM Repo: [Mesh-LLM/mesh-llm](https://github.com/Mesh-LLM/mesh-llm) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Use Mesh LLM's agent launchers so coding agents run on models served by your own machines. **Who it is for:** A developer who wants a coding agent backed by local or pooled models instead of a paid API. **The problem:** Wiring a coding agent to a self-hosted model means finding the right base URL, model name and flags. **The outcome:** Your chosen coding agent talking to the mesh's OpenAI-compatible API. ### Steps 1. Start a node with `mesh-llm serve --auto` or your private mesh 2. Launch the agent through the mesh with `mesh-llm opencode`, `mesh-llm goose`, `mesh-llm claude` or `mesh-llm pi` 3. Pick a model the mesh serves and give the agent a small task first 4. Read the AGENTS guide in the repo for per agent settings ### Prompt to paste into your agent ```text Start Mesh-LLM/mesh-llm and launch through it, choose the best model the mesh serves for coding, and run a small test task in . ``` ### Caveat This card is written from the Mesh-LLM/mesh-llm README, not from a test run by Deduped. Before you start, make sure you have macOS, Linux or Windows. Check the repo for newer instructions before you rely on these steps. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/mesh-llm-mesh-llm/point-claude-code-or-opencode-at-your-mesh/