Brandon Beiler

Definitely hoping to see this feature appear inside Vue.Draggable! Any thoughts on how to hack the current version to makeshift this feature?

@TheGreatRefrigerator Thanks so much for the clarification! I had been poking around in the source and noticed the reference to "swagger-ui.html" and figured I was just not looking in the...

Is there any way to contribute to this MR or body of work in a meaningful way? Would love to help out, but don't want to get in the way.

Amazing! Will watch for that, and see if there is anything in the checklist that I could take a stab at. Thanks!

> I removed the merge commit from the top of your branch so it only contains your additions now - looks good so far > > may not be intended...

@karussell Hey! From playing around with the `Crossing` value, I found it got close but 2 issues 1. It didn't seem to include Stop Signs, though I could be totally...

@Isotr0py Generated an FP8 quant of internvl3-8B to test this out: https://huggingface.co/brandonbeiler/InternVL3-8B-FP8-Dynamic Loaded into Transformers AutoModel with ``` model = AutoModel.from_pretrained( brandonbeiler/InternVL3-8B-FP8-Dynamic, device_map="balanced", # Distribute more evenly across all 4...

Also, worth noting that InternVL3's image -> language mapping layer (mlp1) is skipped in quantizing as the vLLM impl doesn't support scale values because of the primitives used ``` def...

@Isotr0py Awesome! Thanks for the find! Will update across the quantized models and re-upload. Looking closer at the llm_config section between the original model and quantized models (via llm-compressor), it...

Ahh, that makes a lot of sense, especially considering the bnb models experiencing the same missing fields! Will play around locally loading the original model into transformers then saving out...