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
..
chainner_models
frame_interpolation_models
nodes_ace.py
nodes_advanced_samplers.py
nodes_align_your_steps.py
nodes_apg.py
nodes_attention_multiply.py
nodes_audio_encoder.py
nodes_audio.py
nodes_camera_trajectory.py
nodes_canny.py
nodes_cfg.py
nodes_chroma_radiance.py
nodes_clip_sdxl.py
nodes_color.py
nodes_compositing.py
nodes_cond.py
nodes_context_windows.py
nodes_controlnet.py
nodes_cosmos.py
nodes_curve.py
nodes_custom_sampler.py
nodes_dataset.py
nodes_differential_diffusion.py
nodes_easycache.py
nodes_edit_model.py
nodes_eps.py
nodes_flux.py
nodes_frame_interpolation.py
nodes_freelunch.py
nodes_fresca.py
nodes_gits.py
nodes_glsl.py
nodes_hidream.py
nodes_hooks.py
nodes_hunyuan3d.py	Change the save 3d model node's filename prefix to 3d/ComfyUI (CORE-106) (#12826)	2026-04-28 00:59:59 -07:00
nodes_hunyuan.py
nodes_hypernetwork.py
nodes_hypertile.py
nodes_image_compare.py
nodes_images.py
nodes_ip2p.py
nodes_kandinsky5.py
nodes_latent.py
nodes_load_3d.py
nodes_logic.py
nodes_lora_debug.py
nodes_lora_extract.py
nodes_lotus.py
nodes_lt_audio.py
nodes_lt_upsampler.py
nodes_lt.py
nodes_lumina2.py
nodes_mahiro.py
nodes_mask.py	Fix SolidMask and MaskComposite device mismatch with --gpu-only (#13296)	2026-04-28 01:18:19 -07:00
nodes_math.py
nodes_mochi.py
nodes_model_advanced.py
nodes_model_downscale.py
nodes_model_merging_model_specific.py
nodes_model_merging.py
nodes_model_patch.py
nodes_morphology.py
nodes_nag.py
nodes_nop.py
nodes_number_convert.py
nodes_optimalsteps.py
nodes_pag.py
nodes_painter.py
nodes_perpneg.py
nodes_photomaker.py
nodes_pixart.py
nodes_post_processing.py
nodes_preview_any.py	Print more tensor values in the preview any node. (#13544)	2026-04-23 22:19:00 -04:00
nodes_primitive.py
nodes_qwen.py
nodes_rebatch.py
nodes_replacements.py
nodes_resolution.py
nodes_rope.py
nodes_rtdetr.py
nodes_sag.py
nodes_sam3.py
nodes_sd3.py	Make EmptySD3LatentImage node use intermediate dtype. (#13577)	2026-04-26 23:23:57 -04:00
nodes_sdpose.py	SDPose: resize fix (#13656)	2026-05-01 14:17:25 -07:00
nodes_sdupscale.py
nodes_slg.py
nodes_stable3d.py
nodes_stable_cascade.py
nodes_string.py
nodes_tcfg.py
nodes_textgen.py	feat: Gemma4 text generation support (CORE-30) (#13376)	2026-05-02 22:46:15 -04:00
nodes_tomesd.py
nodes_toolkit.py
nodes_torch_compile.py
nodes_train.py
nodes_upscale_model.py
nodes_video_model.py
nodes_video.py
nodes_wan.py
nodes_wanmove.py
nodes_webcam.py
nodes_zimage.py