Overview

Fresh

14

Stale

0

Expired

0

Merged

6
Pull Requests

fix(qwen3_vl): size video budget for the padded frame count in smart_resize

Remove now useless keys_to_ignore_at_inference in all configs

Simplify cache handling in generate

Set _skip_keys_device_placement directly inside PreTrainedModel

Load more

Commit 40ed19f

Gather —

Logic —

Template —