We need to migrate over a set of models to s3 fom their new Red Hat AI stubs. This set of models are not currently in S3, and they are the ones we still need to move to S3 (for the RHOAI catalog demo):
Qwen2.5-VL-72B-Instruct-quantized.w8a8
Pixtral-Large-Instruct-2411-hf-quantized.w8a8
whisper-large-v2-W4A16-G128
Llama-4-Scout-17B-16E-Instruct-FP8-dynamic

We're still waiting on quantized mistral 3.1, but I will follow up w/ the research team and comment on this ticket when we have that added to the HF and need to pull it into S3.