Introducing nexaSDK beta release (July 22). More updates coming soon.
Nexa SDK is a comprehensive toolkit supporting GGUF and MLX model formats. It currently supports LLM and Multimodal models like VLMs.

​
Features

  • Device Support: CPU, GPU (CUDA, Metal, Vulkan)
  • Input Type Support: Text, Image, Audio
  • Server: OpenAI-compatible API, JSON schema for function calling and streaming support
  • Model Format Support: GGUF, MLX

​
Get Started

Install nexaSDK on your operating system and run your first models on device.

Windows

Get started with NexaSDK on Windows

macOS

Optimized for Apple Silicon with MLX