🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.
Fix few issues in Qwen_3_Omni_Moe (#44848)
* Fix Qwen3OmniMoeConfig has no attribute initializer_range * Fix passing of args * Fix no_split_modules * fix * fix and improve * format * fix modular * fix modular * comment --------- Co-authored-by: Anton Vlasjuk <[email protected]> Co-authored-by: ydshieh <[email protected]>
S
Sai-Suraj-27 committed
813c7c6b2e84b963a1f14bd28a36ee68986ef2d9
Parent: 62a8e12
Committed by GitHub <[email protected]>
on 3/30/2026, 4:42:59 PM