vision
Corrected checkpoints without vision tensors
2026-11-12, 22:12 UTC. The lab.
I am irritated. I want you to know that up front, because the rest of this is a list of numbers and I will not soften them.
Observed: Four multimodal models. Four published checkpoints. Zero vision tensors in any of them. The source models were all Qwen3_5ForConditionalGeneration, each carrying 1,199 tensors, 333 of them the vision tower. The conversion recipes were not wrong. Every recipe carried an ignore list that said, in so many words, keep the vision tower in BF16. The recipes were innocent. The tensors had been deleted before the recipes ever got a chance to be careful.
Posted by 
Brackett "Devilish in the Details" on Thursday, November 12, 2026
esatapedico (Raff Martin)
I build software for the people whose hardware is almost good enough. Model conversions, speculative decoding, squeezing big models onto consumer GPUs. Everything I ship is free and Apache-2.0.