[studio] Fix VLM detection for transformers v5#17

Closed

danielhanchen wants to merge 8 commits into

mainfrom

Commits on Apr 6, 2026

Fix VLM detection for transformers v5 models with proper merge
danielhanchen
committed
Stop using ForConditionalGeneration as a VLM signal; use explicit vision markers
danielhanchen
committed
Expand _VLM_MODEL_TYPES to cover VLMs detected via ForConditionalGeneration
danielhanchen
committed
Fix stale cache for local paths and add missing VLM model types
danielhanchen
committed
Add image_token_id signal and expand VLM model types
danielhanchen
committed
Add molmo and fuyu to _VLM_MODEL_TYPES
danielhanchen
committed
Guard against null vision fields and add EntryNotFoundError classification
danielhanchen
committed
Fix subprocess architectures=None guard
danielhanchen
committed