Neural Compressor Model Hardware Deployment and Compatibility #2086
Description
Aims to clarify the HW deployment capabilities of Neural Compressor optimized models.
- Specifically I'd like to know if these models are optimized for specific architectures such as x86 or if they are compatible with a wide range of HW (such as ARM) platforms.
- Are there any known issues or limitations when deploying optimized models on non-x86 architectures such as ARM, RISC-V, or others?
- What are Deployment Considerations for Non-x86 Platforms ?
- Understanding the HW compatibility will help assess the potential deployment scenarios for our models.