Add support for phi3-vision

intel / neural-speed

An innovative library for efficient LLM inference via low-bit quantization

https://github.com/intel/neural-speed

Apache License 2.0

345 stars 37 forks source link

Add support for phi3-vision #268

Open bil-ash opened 4 months ago

bil-ash commented 4 months ago

Please add support for phi3-vision to neural speed. According to benchmarks provided, it comes close to many SOTA multimodal models with only a fraction of the size - which is perfect for running on client CPUs and intel GPUs.

intellinjun commented 4 months ago

@bil-ash Thank you for your suggestion, we will assess the needs internally and get back to you as soon as possible.