
@0xSero I am very curious - how hard would it be to add vision support to GLM5.2 model directly? like I don't mean that it would use tool call to call VLM model but like integrate it to that model. I truly believe it is huge missed opportunity that GLM5.2 is not multimodal :/
English








