Jukka Seppänen be95871adc feat: Gemma4 text generation support (CORE-30) (#13376) ...

* initial gemma4 support

* parity with reference implementation

outputs can 100% match transformers with same sdpa flags, checkpoint this and then optimize

* Cleanup, video fixes

* cleanup, enable fused rms norm by default

* update comment

* Cleanup

* Update sd.py

* Various fixes

* Add fp8 scaled embedding support

* small fixes

* Translate think tokens

* Fix image encoder attention mask type

So it works with basic attention

* Handle thinking tokens different only for Gemma4

* Code cleanup

* Update nodes_textgen.py

* Use embed scale class instead of buffer

Slight difference to HF, but technically more accurate and simpler code

* Default to fused rms_norm

* Update gemma4.py

2026-05-02 22:46:15 -04:00

..

diffusionmodules

feat: SUPIR model support (CORE-17) (#13250)

2026-04-18 23:02:01 -04:00

distributions

Small optimizations.

2024-12-18 18:23:28 -05:00

encoders

Remove dead code. (#13251)

2026-04-01 20:22:06 -04:00

attention.py

feat: Gemma4 text generation support (CORE-30) (#13376)

2026-05-02 22:46:15 -04:00

ema.py

New Year ruff cleanup. (#11595)

2026-01-01 22:06:14 -05:00

sdpose.py

SDPose: resize input always (#13349)

2026-04-10 11:26:55 -10:00

sub_quadratic_attention.py

Widen OOM_EXCEPTION to AcceleratorError form (#12835)

2026-03-10 00:41:02 -04:00

temporal_ae.py

Basic Hunyuan Video model support.

2024-12-16 19:35:40 -05:00