Steven Leve

Thanks for sharing this research, I'm looking forward to exploring the notebook.

I ran some experiments with OpenVINO on my much more modest I7 with Intel Xe GPU & 32GB DDR4 - a much more limited system amd common business laptop (Dell Latitude 5420). My experiment wasn't formal, but used the same text prompts for consistency.

Interestingly I found that the OpenVINO sped up Mistral 7b but Llama 3.1 8B ran slower. In most cases the GPU lift was marginal on this hardware. Qwen3-VL 8B was the newest model I tried and it ran the fastest - only surpassed by Qwen3 3B.

Benchmarking with OpenVINO
https://www.linkedin.com/pulse/benchmarking-openvino-steven-leve-gcg7c?utm_source=share&utm_medium=member_android&utm_campaign=share_via