Harry Mellor 45273b8dcb [1/N] Harden Transformers modelling backend multi-modal path (#51408) ...

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

2026-08-07 17:30:43 +00:00

..

generation

[MM][CG] Support ViT full CUDA graph for Ernie-4.5-VL image inference (#45254)

2026-08-06 05:04:19 +00:00

generation_ppl_test

[Model] Fused mm preprocess normalisation on the Device (#50411)

2026-08-06 03:11:14 -07:00

pooling

[Bugfix] Respect declared attention contract for ColQwen3.5 retrievers (#49372)

2026-07-26 04:08:07 +00:00

processing

[1/N] Harden Transformers modelling backend multi-modal path (#51408)

2026-08-07 17:30:43 +00:00

__init__.py

[CI/Build] Move model-specific multi-modal processing tests (#11934)

2025-01-11 13:50:05 +08:00

conftest.py

Deprecated 1st generation Qwen and QwenVL models (#45131)

2026-06-10 14:55:33 +00:00

test_mapping.py

[bugfix] Fix Cosmos3 Edge checkpoint weights filtering, video loading, prompt expansion (#49190)

2026-07-21 17:18:36 +08:00

test_nano_nemotron_vl.py

Avoid silent weights corruption when loading Nemotron Nano VL with reusable-buffer loaders like runai distributed streaming (#42244)

2026-05-11 12:03:14 +00:00