Ishrat Badami

icon indicating a website link https://triplelift.com/

icon company Triplelift icon location Ottawa, Canada icon location CTV Computer Vision and Machine Learning Engineer

Results 5 comments of Ishrat Badami

assert token_byte_positions[-1] == last_pos on gen()

Facing the same problem with models.Transformers mistral models.

image is broken for LLaVa models

I am facing this same problem. Any update on this issue?

image is broken for LLaVa models

I don't think supporting images for opensource multi-modal models is pursued actively, https://github.com/guidance-ai/guidance/issues/554#issuecomment-1878163908 This support was introduced only for Gemini models.

image is broken for LLaVa models

> I have been digging into this a bit. You can pass a Llama model directly into guidance's models.LlamaCpp and use it. However: the `image` object that guidance uses seems...

image is broken for LLaVa models

@slundberg Could you please provide us with high-level pointers to start our dev efforts? Thx!