This is an AI code assistant powered by multiple backends including Cerebras and OpenRouter, supporting various models including Llama 3.3 70B and Qwen 2.5 32B.

Inspired by Hassan's Llama Coder.

• Multiple Backends: Choose between Cerebras (fast inference) and OpenRouter (more model options)
• Multiple Models:
  • Cerebras: Llama 3.3 70B, Llama 3.1 70B, Llama 3.1 8B
  • OpenRouter: Qwen 2.5 32B, Llama 3.1 70B/8B, Claude 3.5 Sonnet, GPT-4o
• Settings Panel: Easy switching between backends and models
• Real-time Code Generation: Generate complete HTML applications instantly

1. Sign up for Cerebras
2. Get a Cerebras API Key
3. Save it in your project env variable called CEREBRAS_API_KEY

1. Sign up for OpenRouter
2. Get an OpenRouter API Key
3. Save it in your project env variable called OPENROUTER_API_KEY

You can use either backend independently or set up both for maximum flexibility.