mlx-vlm
mlx-vlm copied to clipboard
Qwen2-VL: Model Test Structure Improvements
With the introduction of Qwen2-VL and growing model support, the testing framework could be refactored to maintain better separation of functionality between models
To start, I'll address the request from https://github.com/Blaizzy/mlx-vlm/pull/63#issuecomment-2386775484 separating out the Qwen2-VL tests into their own specific model / folder.
@benzimring that's perfect.
Please do!
I have been a bit busy with v0.1.0 and work but I'm back.
Could you send a PR with your proposed changes?