🤝 Supported chipmakers 
      
        
        
        
      
    
  


    


    



    


    

NexaSDK - Run any AI model on any backend

NexaSDK is an easy-to-use developer toolkit for running any AI model locally — across NPUs, GPUs, and CPUs — powered by our NexaML engine, built entirely from scratch for peak performance on every hardware stack. Unlike wrappers that depend on existing runtimes, NexaML is a unified inference engine built at the kernel level. It’s what lets NexaSDK achieve Day-0 support for new model architectures (LLMs, multimodal, audio, vision). NexaML supports 3 model formats: GGUF, MLX, and Nexa AI's own .nexa format.

⚙️ Differentiation

  Legend:
  ✅ Supported   |  
  ⚠️ Partial or limited support    |  
  ❌ No

Recent Wins

• 📣 Release Nexa AI’s AutoNeural-VL-1.5B, an NPU-native vision–language model built for real-time in-car assistants, delivering 14× lower latency, 3× faster decode, and 4× longer context on Qualcomm SA8295P — now also runnable on Qualcomm X Elite laptops.
• 📣 Support Mistral AI's Ministral-3-3B across Qualcomm Hexagon NPU, Apple Neural Engine, GPU and CPU.
• 📣 Release Linux SDK for NPU/GPU/CPU. See Linux SDK Doc (https://docs.nexa.ai/nexa-sdk-docker/overview).
• 📣 Support Apple Neural Engine for Granite-4.0 (https://huggingface.co/NexaAI/Granite-4-Micro-ANE), Qwen3 (https://huggingface.co/NexaAI/Qwen3-0.6B-ANE), Gemma3 (https://huggingface.co/NexaAI/Gemma3-1B-ANE), and Parakeetv3 (https://huggingface.co/NexaAI/parakeet-tdt-0.6b-v3-ane). Download NexaSDK for ANE here (https://nexa-model-hub-bucket.s3.us-west-1.amazonaws.com/public/nexa_sdk/downloads/nexa-cli_macos_arm64_ane.pkg).
• 📣 Support Android SDK for NPU/GPU/CPU. See Android SDK Doc (https://docs.nexa.ai/nexa-sdk-android/overview) and Android SDK Demo App.
• 📣 Support SDXL-turbo image generation on AMD NPU. See AMD blog : Advancing AI with Nexa AI (https://www.amd.com/en/developer/resources/technical-articles/2025/advancing-ai-with-nexa-ai--image-generation-on-amd-npu-with-sdxl.html).
• Support Android Python SDK for NPU/GPU/CPU. See Android Python SDK Doc (https://docs.nexa.ai/nexa-sdk-android/python) and Android Python SDK Demo App.
• 📣 Day-0 Support for Qwen3-VL-4B and 8B in GGUF, MLX, .nexa format for NPU/GPU/CPU. We are the only framework that supports the GGUF format. Featured in Qwen's post about our partnership (https://x.com/Alibaba_Qwen/status/1978154384098754943).
• 📣 Day-0 Support for IBM Granite 4.0 on NPU/GPU/CPU. NexaML engine were featured right next to vLLM, llama.cpp, and MLX in IBM's blog (https://x.com/IBM/status/1978154384098754943).
• 📣 Day-0 Support for Google EmbeddingGemma on NPU. We are featured in Google's social post (https://x.com/googleaidevs/status/1969188152049889511).
• 📣 Supported vision capability for Gemma3n: First-ever Gemma-3n (https://sdk.nexa.ai/model/Gemma3n-E4B) multimodal inference for GPU & CPU, in GGUF format.
• 📣 Intel NPU Support DeepSeek-r1-distill-Qwen-1.5B (https://sdk.nexa.ai/model/DeepSeek-R1-Distill-Qwen-1.5B-Intel-NPU) and Llama3.2-3B (https://sdk.nexa.ai/model/Llama3.2-3B-Intel-NPU)
• 📣 Apple Neural Engine Support for real-time speech recognition with Parakeet v3 model (https://sdk.nexa.ai/model/parakeet-v3-ane)