Skip to content

v1.0.7 — OpenAI Compatible API, Ollama, Multilingual Support

Latest
Latest

Choose a tag to compare

View all tags
@notadev-iamaura notadev-iamaura released this 07 Mar 15:30
· 302 commits to main since this release
v1.0.7
6ce5836

Highlights

OpenAI Compatible API (/v1/chat/completions)

  • Connect existing OpenAI SDK code directly to OneRAG — no code changes needed
  • Works with LangChain, Cursor, Open WebUI, and any OpenAI-compatible tool
  • Streaming (SSE) support with "stream": true
  • Model selection via model field: gemini, ollama/qwen2.5:3b, openrouter/google/gemini-2.0-flash

Ollama Local LLM Integration

  • Run completely offline with local LLM — no API key required
  • Auto-detection: if Ollama is running, it's used automatically
  • ollama pull llama3.2 && make easy-start — that's it

Multilingual Easy-Start (4 Languages)

  • make easy-start LANG=en (English), LANG=ja (日本語), LANG=zh (中文), default: 한국어
  • UI text, system prompts, and sample data fully localized

Other Improvements

  • 2,100+ tests passing
  • CLI demo animation added to README
  • Architecture SVG diagram
  • Enhanced PyPI keywords and GitHub Topics for discoverability

Quick Start

git clone https://github.com/notadev-iamaura/OneRAG.git && cd OneRAG && uv sync

# Full API Server (Docker)
cp quickstart/.env.quickstart .env && make start

# Local CLI Chatbot (No Docker)
make easy-start

# Fully offline with Ollama
ollama pull llama3.2 && make easy-start

Full Changelog: v1.2.1...v1.0.7

Assets 2
Loading