(suggestion) https://huggingface.co/baseten/GLM-5.2-Vision-FP8
https://huggingface.co/baseten/GLM-5.2-Vision-FP8
I think that I might come across as a bit of noisy post. (Sorry)
But I鈥榙 like to make a suggestion regarding the quantization of this model for your consideration.
My understanding is that this model simply adds a vision encoder and a projector and updates the chat template, without any additional training of the language model itself. Therefore, if you swap the chat template (image add) of the existing GLM-5.2-gguf, you could reduce the conversion effort to just adding the GGUF files for the vision encoder and projector.
we want a proper quantization, not a simple hack, plus not sure how to do it as chat template is usually embedded within the gguf model itself, plus we need to do imatrix one way or another. Plus we would then need to depend on already quantized glm-5.2, which... is also not done, so we would need to quant both