Byte

Byte is a desktop application that runs local AI models like Llama and Mistral directly on the user's machine, eliminating the need for cloud dependency and subscription fees. It also supports bring-your-own-key for cloud models from providers like OpenAI, Claude, and Gemini, allowing users to pay providers directly at published rates. The app targets developers, IT professionals, and individual users who want privacy, cost savings, and flexibility in AI model usage. It is positioned as a free, open-source alternative to subscription-based AI chat apps, with no markup or tracking.

Key features

• Runs local models like Llama and Mistral
• Supports 16 AI providers with 30+ models
• Bring your own key for cloud models
• No markup on cloud provider rates
• Open source and free to download
• Unlimited conversations
• Full model control
• Built-in memory
• Switch models mid-conversation
• Web search support
• No tracking or telemetry
• Data stays on device