https://github.com/spartan-71/helix
A terminal-native AI chat interface — talk to any LLM, right from your shell.
https://github.com/spartan-71/helix
ai claude gemma gpt llama llm mixtral-8x7b ollama ollama-client python terminal tui
Last synced: 4 months ago
JSON representation
A terminal-native AI chat interface — talk to any LLM, right from your shell.
- Host: GitHub
- URL: https://github.com/spartan-71/helix
- Owner: Spartan-71
- License: apache-2.0
- Created: 2025-06-09T07:30:48.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-06-11T17:58:29.000Z (about 1 year ago)
- Last Synced: 2025-06-16T11:52:30.153Z (12 months ago)
- Topics: ai, claude, gemma, gpt, llama, llm, mixtral-8x7b, ollama, ollama-client, python, terminal, tui
- Language: Python
- Homepage: https://pypi.org/project/helix-chat/
- Size: 145 KB
- Stars: 6
- Watchers: 0
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
A terminal-native AI chat interface — talk to any LLM, right from your shell.
---
Helix is a work-in-progress project that brings the power of modern LLMs to your command line. Whether you prefer local models like LLaMA 3 or cloud-based ones like GPT-4 or Claude, Helix is being built to give you a ChatGPT-like experience with zero distractions and maximum speed — right inside your terminal.
---
## 🚧 Status
Helix is currently under **rapid development** and not yet production-ready.
Stay tuned for the first beta release — coming soon!
## ✅ Planned Features
- [x] Terminal-native Interface
- [ ] Chat with any LLM
- [ ] API-based (OpenAI, Claude, Gemini, etc.)
- [ ] Local models (LLaMA 3, Mistral, Phi-3, Gemma, etc.) via Ollama/LlamaEdge
- [x] Unified backend using **LiteLLM**
- [ ] Seamless Model Switching
- [ ] RAG for Context-Aware Chat
- [ ] Persistent Chat History + Fuzzy Search
- [ ] Multi-modal Input Support (Text + Image)
- [x] Markdown Rendering in Terminal
- [ ] Streaming Token-by-Token Response Output
- [ ] Theming Support (Dark, Solarized, etc.)
- [ ] Offline Mode Toggle
- [ ] Support for MCP Servers
- [ ] Slash Commands (e.g. `/reset`, `/save`, `/switch`)
- [ ] Keyboard Shortcuts (Vim-style navigation)
- [ ] Export Chats (Markdown, JSON, HTML)
## 💡 Why Helix?
Just like the double helix structure of DNA, this project weaves together the raw power of terminals and the evolving intelligence of LLMs — forming a beautifully minimal, developer-first chat experience.
## 📦 Installation
Download: `pip install helix-chat`
Launch: `helix`
>Make sure you have ollama/llama3 model pulled locally, since helix uses that by default.
## 🤝 Contributing
We’ll soon open up issues and contribution guidelines. If you're interested in helping, feel free to ⭐ the repo and watch for updates!
## 📜 License
Apache License 2.0 — Free to use, modify, and distribute, with proper attribution and a focus on contribution transparency.
You can read the full license [here](LICENSE).