---
title: P2p Llm
emoji: 🚀
colorFrom: green
colorTo: purple
sdk: docker
sdk_version: 5.13.0
pinned: false
license: agpl-3.0
short_description: peer to peer LLM inference
app_file: app.py
app_port: 8080
models:
- deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
- unsloth/DeepSeek-R1-Distill-Qwen-1.5B-GGUF
---

# P2P LLM Inference Platform

A peer-to-peer platform for distributed LLM inference.

## Current Features

- **Trusted Peers Management**:
  - Load peers from configuration file
  - Load peers from database
  - Combine both sources for peer discovery
- **Request Forwarding**:
  - Distribute requests across available peers
  - Handle both regular and SSE responses

## Planned Features

- **Trust Scores**:
  - Track peer reliability and performance
  - Weight request distribution based on scores
- **Peer Advertising**:
  - Automatic peer discovery
  - Peer-to-peer network formation
- **Enhanced Security**:
  - Peer authentication
  - Request validation

## Getting Started

1. Clone the repository
2. Configure `files/config.json` with your settings
3. Build and run with Docker:

```bash
docker-compose up --build
```

## Configuration

Edit `files/config.json` to specify:

- Database path
- Target URL
- Trusted peers (URLs and public keys)
- Trusted peers file path

## Development

```bash
# Run locally
go run main.go

# Run tests
go test ./...
```

Check out the configuration reference at <https://huggingface.co/docs/hub/spaces-config-reference>