Does it support FastVLM-0.5B-4bit version?

[Lornatang]

I used the mlx-vlm tool to perform 4-bit quantization on the official FastVLM-0.5B and modified it according to your example, but I encountered this error：

MLXNN/Module.swift:570: Fatal error: 'try!' expression unexpectedly raised an error: MLXNN.UpdateError.needModuleInfo("Unable to get @ModuleInfo for FastVLMMultiModalProjector.layers -- must be wrapped to receive updates")

It would be great if you could take a look at this. Thank you for your great work.

[davidkoski]

Which FastVLM are you referring to?

• https://github.com/apple/ml-fastvlm
• https://github.com/ml-explore/mlx-swift-lm (https://github.com/ml-explore/mlx-swift-lm/blob/main/Libraries/MLXVLM/Models/FastVLM.swift)

I suspect the latter as FastVLMMultiModalProjector there is-a Sequential and has a layers property.

[davidkoski]

Ah yeah, that wouldn't support quantization as it is a let property.

Let's move this to mlx-swift