Qwen3-8B-NPU

Qwen3-8B

Run Qwen3-8B optimized for Qualcomm Hexagon NPUs with NexaSDK.

Quickstart

1. Install NexaSDK

2. Activate your device with your access token for free at sdk.nexa.ai

   nexa config set license '<access_token>'
   

3. Run the model on Qualcomm NPU in one line:

   nexa infer NexaAI/Qwen3-8B-NPU
   

Model Description

Qwen3-8B is an 8-billion-parameter large language model developed by the Qwen team at Alibaba Cloud.
It sits in the mid-range of the Qwen3 family, offering a strong balance between capability and efficiency — ideal for both enterprise-scale applications and high-performance local inference.

Trained on a vast multilingual and multimodal corpus, Qwen3-8B demonstrates advanced performance in reasoning, dialogue, and code generation while maintaining efficient inference on modern GPUs and NPUs.

Features

• High reasoning ability: excels at logic, math, and structured problem solving.
• Robust dialogue understanding: context-aware multi-turn conversation with memory.
• Creative and analytical writing: supports long-form generation and precise summarization.
• Multilingual support: strong understanding and generation across major global languages.
• Fine-tunable: adaptable for specialized enterprise and domain applications.

Use Cases

• Advanced conversational AI and digital assistants
• Multilingual enterprise documentation and summarization
• Coding copilots and software reasoning agents
• Data analysis and business intelligence automation
• Education, research, and domain-specific tutoring

Inputs and Outputs

Input:

• Natural language or structured text prompts.
• Optional multimodal inputs (text + vision in compatible variants).

Output:

• Natural language generation (answers, code, analysis, creative writing).
• Structured data outputs or logits for downstream integration.

License

This model and repository are released under the Creative Commons Attribution–NonCommercial 4.0 (CC BY-NC 4.0) license.
Use and modification are permitted for non-commercial purposes with proper attribution.
For commercial licensing or enterprise inquiries, contact [email protected].