Repolode
Submit
yomorun

yomorun/yomo

An open-source LLM function calling framework in Rust for deploying serverless AI tools and agents across geo-distributed edge infrastructure.

View on GitHubWebsite
Stars
1,930
Forks
150
Open issues
1
Last commit
13 hours ago

What is it?

YoMo is a framework for building AI agents that call LLM tools (functions or skills). Developers write tool logic as serverless functions, and YoMo handles deploying, running and managing them behind an OpenAI-compatible chat completions endpoint. A typical setup uses Ollama as the model provider, with the CLI starting a server and running tools from a project scaffolded by `yomo init`.

The project focuses on latency for AI interactions. Communication between components is encrypted with TLS 1.3, and the architecture is designed to place inference and tools close to end users rather than in a single data center. Tools can also be invoked directly over HTTP, without going through the LLM.

The repository is written in Rust and installs through a shell script that provides the `yomo` CLI. Example projects are published separately in the llm-function-calling-examples repository.

Key features

  • 01Deploys and manages LLM tools as serverless functions.
  • 02Encrypts data packets with TLS 1.3.
  • 03Exposes an OpenAI-compatible chat completions endpoint.
  • 04Supports Ollama as a local LLM provider.
  • 05Distributes inference and tools across geo-distributed edge locations.
  • 06Provides a CLI for serving, scaffolding and running tools.

Use cases

  • Building a chatbot that calls external APIs such as weather lookups through function calling.
  • Serving latency-sensitive AI agents to users in several regions.
  • Packaging and iterating on LLM tools or skills without managing servers for each one.

Quick start

  1. 1

    Install the YoMo CLI with the official install script

  2. 2

    Pull the ornith model for Ollama

    ollama pull ornith
  3. 3

    Start the YoMo server

    yomo serve
  4. 4

    Scaffold a project and run the get-weather tool

    yomo init

Things to know

  • The quick start expects a local Ollama LLM provider with a downloaded model.

In collections

Claude Code tools, skills & pluginsMCP servers

Topics

#ai#llm#function-calling#agent-framework#serverless#edge-computing#rust#mcp#a2a-protocol#claude-code

FAQ

What is yomo?
An open-source LLM function calling framework in Rust for deploying serverless AI tools and agents across geo-distributed edge infrastructure. YoMo is a framework for building AI agents that call LLM tools (functions or skills). Developers write tool logic as serverless functions, and YoMo handles deploying, running and managing them behind an OpenAI-compatible chat completions endpoint. A typical setup uses Ollama as the model provider, with the CLI starting a server and running tools from a project scaffolded by `yomo init`.
Is yomo free to use? What license does it have?
yomo has no declared license. Its code is public, but without a license you have no explicit permission to reuse it; ask the author before relying on it.
Is yomo still maintained?
Actively maintained. The last commit to yomo was on Nov 7, 2026.
What can I use yomo for?
Building a chatbot that calls external APIs such as weather lookups through function calling. Serving latency-sensitive AI agents to users in several regions. Packaging and iterating on LLM tools or skills without managing servers for each one.
How do I install yomo?
Install the YoMo CLI with the official install script Pull the ornith model for Ollama: ollama pull ornith Start the YoMo server: yomo serve Scaffold a project and run the get-weather tool: yomo init
Where can I get yomo?
The source code and installation instructions are on GitHub: https://github.com/yomorun/yomo

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects