Aryan 7b904941bc Cosmos (#10660) ...

* begin transformer conversion

* refactor

* refactor

* refactor

* refactor

* refactor

* refactor

* update

* add conversion script

* add pipeline

* make fix-copies

* remove einops

* update docs

* gradient checkpointing

* add transformer test

* update

* debug

* remove prints

* match sigmas

* add vae pt. 1

* finish CV* vae

* update

* update

* update

* update

* update

* update

* make fix-copies

* update

* make fix-copies

* fix

* update

* update

* make fix-copies

* update

* update tests

* handle device and dtype for safety checker; required in latest diffusers

* remove enable_gqa and use repeat_interleave instead

* enforce safety checker; use dummy checker in fast tests

* add review suggestion for ONNX export

Co-Authored-By: Asfiya Baig <asfiyab@nvidia.com>

* fix safety_checker issues when not passed explicitly

We could either do what's done in this commit, or update the Cosmos examples to explicitly pass the safety checker

* use cosmos guardrail package

* auto format docs

* update conversion script to support 14B models

* update name CosmosPipeline -> CosmosTextToWorldPipeline

* update docs

* fix docs

* fix group offload test failing for vae

---------

Co-authored-by: Asfiya Baig <asfiyab@nvidia.com>

2025-05-07 20:59:09 +05:30

..

__init__.py

Fix conversion script

2022-07-15 17:00:41 +00:00

change_naming_configs_and_checkpoints.py

[chore] change licensing to 2025 from 2024. (#10615)

2025-01-20 16:57:27 -10:00

conversion_ldm_uncond.py

[OmegaConf] replace it with yaml (#6488)

2024-01-15 20:02:10 +05:30

convert_amused.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_animatediff_motion_lora_to_diffusers.py

[core] AnimateDiff SparseCtrl (#8897)

2024-07-26 17:46:05 +05:30

convert_animatediff_motion_module_to_diffusers.py

[Pipeline] AnimateDiff SDXL (#6721)

2024-05-08 21:27:14 +05:30

convert_animatediff_sparsectrl_to_diffusers.py

[core] AnimateDiff SparseCtrl (#8897)

2024-07-26 17:46:05 +05:30

convert_asymmetric_vqgan_to_diffusers.py

Asymmetric vqgan (#3956)

2023-07-20 17:51:06 +02:00

convert_aura_flow_to_diffusers.py

[Core] Add AuraFlow (#8796)

2024-07-11 08:50:19 -10:00

convert_blipdiffusion_to_diffusers.py

Fix style (#10478)

2025-01-07 11:06:36 +05:30

convert_cogvideox_to_diffusers.py

CogVideoX 1.5 (#9877)

2024-11-19 00:56:34 +05:30

convert_cogview3_to_diffusers.py

[Fix] Syntax error (#10068)

2024-12-02 11:28:00 +05:30

convert_cogview4_to_diffusers_megatron.py

CogView4 Control Block (#10809)

2025-03-15 07:15:56 -10:00

convert_cogview4_to_diffusers.py

CogView4 Control Block (#10809)

2025-03-15 07:15:56 -10:00

convert_consistency_decoder.py

chore: remove redundant words (#10609)

2025-01-20 08:15:26 -10:00

convert_consistency_to_diffusers.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_cosmos_to_diffusers.py

Cosmos (#10660)

2025-05-07 20:59:09 +05:30

convert_dance_diffusion_to_diffusers.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_dcae_to_diffusers.py

[DC-AE] Add the official Deep Compression Autoencoder code(32x,64x,128x compression ratio); (#9708)

2024-12-07 01:01:51 +05:30

convert_ddpm_original_checkpoint_to_diffusers.py

Ruff: apply same rules as in transformers (#2827)

2023-03-27 16:18:57 +02:00

convert_diffusers_sdxl_lora_to_webui.py

changed positional parameters to named parameters like in docs (#6905)

2024-02-08 21:39:03 +05:30

convert_diffusers_to_original_sdxl.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_diffusers_to_original_stable_diffusion.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_dit_to_diffusers.py

Replace flake8 with ruff and update black (#2279)

2023-02-07 23:46:23 +01:00

convert_flux_to_diffusers.py

Fix typos in docs and comments (#11416)

2025-04-30 20:30:53 -10:00

convert_flux_xlabs_ipadapter_to_diffusers.py

Support Flux IP Adapter (#10261)

2024-12-21 17:49:58 +00:00

convert_gligen_to_diffusers.py

Remove torch_dtype in to() to end deprecation (#6886)

2024-02-08 09:38:57 +05:30

convert_hunyuan_video_to_diffusers.py

New HunyuanVideo-I2V (#11066)

2025-03-24 21:18:40 +05:30

convert_hunyuandit_controlnet_to_diffusers.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_hunyuandit_to_diffusers.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_i2vgen_to_diffusers.py

[chore] change licensing to 2025 from 2024. (#10615)

2025-01-20 16:57:27 -10:00

convert_if.py

Update access of configuration attributes (#7343)

2024-03-18 08:53:29 -10:00

convert_k_upscaler_to_diffusers.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_kakao_brain_unclip_to_diffusers.py

[Core] move transformer scripts to transformers modules (#6747)

2024-01-29 22:28:28 +05:30

convert_kandinsky3_unet.py

[@cene555][Kandinsky 3.0] Add Kandinsky 3.0 (#5913)

2023-11-24 17:46:00 +01:00

convert_kandinsky_to_diffusers.py

[Core] move transformer scripts to transformers modules (#6747)

2024-01-29 22:28:28 +05:30

convert_ldm_original_checkpoint_to_diffusers.py

[chore] change licensing to 2025 from 2024. (#10615)

2025-01-20 16:57:27 -10:00

convert_lora_safetensor_to_diffusers.py

[LoRA test suite] refactor the test suite and cleanse it (#7316)

2024-03-20 17:13:52 +05:30

convert_ltx_to_diffusers.py

LTX 0.9.5 (#10968)

2025-03-17 16:43:36 -10:00

convert_lumina_to_diffusers.py

Rename Lumina(2)Text2ImgPipeline -> Lumina(2)Pipeline (#10827)

2025-03-13 09:24:21 -10:00

convert_mochi_to_diffusers.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_models_diffuser_to_diffusers.py

Ruff: apply same rules as in transformers (#2827)

2023-03-27 16:18:57 +02:00

convert_ms_text_to_video_to_diffusers.py

[chore] change licensing to 2025 from 2024. (#10615)

2025-01-20 16:57:27 -10:00

convert_music_spectrogram_to_diffusers.py

#7535 Update FloatTensor type hints to Tensor (#7883)

2024-05-10 09:53:31 -10:00

convert_ncsnpp_original_checkpoint_to_diffusers.py

[chore] change licensing to 2025 from 2024. (#10615)

2025-01-20 16:57:27 -10:00

convert_omnigen_to_diffusers.py

Add OmniGen (#10148)

2025-02-12 02:16:38 +05:30

convert_original_audioldm2_to_diffusers.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_original_audioldm_to_diffusers.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_original_controlnet_to_diffusers.py

[chore] change licensing to 2025 from 2024. (#10615)

2025-01-20 16:57:27 -10:00

convert_original_musicldm_to_diffusers.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_original_stable_diffusion_to_diffusers.py

[chore] change licensing to 2025 from 2024. (#10615)

2025-01-20 16:57:27 -10:00

convert_original_t2i_adapter.py

[chore] change licensing to 2025 from 2024. (#10615)

2025-01-20 16:57:27 -10:00

convert_pixart_alpha_to_diffusers.py

Fix PixArt 256px inference (#6789)

2024-03-03 10:31:21 +05:30

convert_pixart_sigma_to_diffusers.py

PixArt-Sigma Implementation (#7654)

2024-04-23 22:33:08 -10:00

convert_sana_controlnet_to_diffusers.py

[ControlNet] Adds controlnet for SanaTransformer (#11040)

2025-04-13 19:19:39 +05:30

convert_sana_to_diffusers.py

Fix typos in docs and comments (#11416)

2025-04-30 20:30:53 -10:00

convert_sd3_controlnet_to_diffusers.py

Sd35 controlnet (#10020)

2024-11-27 10:44:48 -10:00

convert_sd3_to_diffusers.py

[Fix] Syntax error (#10068)

2024-12-02 11:28:00 +05:30

convert_shap_e_to_diffusers.py

Fix typos in docs and comments (#11416)

2025-04-30 20:30:53 -10:00

convert_stable_audio.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_stable_cascade_lite.py

Update Stable Cascade Conversion Scripts (#7271)

2024-03-13 12:35:44 +05:30

convert_stable_cascade.py

Update Stable Cascade Conversion Scripts (#7271)

2024-03-13 12:35:44 +05:30

convert_stable_diffusion_checkpoint_to_onnx.py

change to 2024 in the license (#6902)

2024-02-08 08:19:31 -10:00

convert_stable_diffusion_controlnet_to_onnx.py

Convert Stable Diffusion ControlNet to TensorRT (#4465)

2023-08-11 08:12:26 +05:30

convert_stable_diffusion_controlnet_to_tensorrt.py

Convert Stable Diffusion ControlNet to TensorRT (#4465)

2023-08-11 08:12:26 +05:30

convert_svd_to_diffusers.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_tiny_autoencoder_to_diffusers.py

Remove code snippets containing is_safetensors_available() (#4521)

2023-08-11 11:05:22 +05:30

convert_unclip_txt2img_to_image_variation.py

Replace flake8 with ruff and update black (#2279)

2023-02-07 23:46:23 +01:00

convert_unidiffuser_to_diffusers.py

[WIP] Refactor UniDiffuser Pipeline and Tests (#4948)

2023-10-02 18:24:55 +02:00

convert_vae_diff_to_onnx.py

make style

2023-03-06 10:40:18 +00:00

convert_vae_pt_to_diffusers.py

[BUG] Fix convert_vae_pt_to_diffusers bug (#11078)

2025-04-10 06:59:45 +01:00

convert_versatile_diffusion_to_diffusers.py

[chore] change licensing to 2025 from 2024. (#10615)

2025-01-20 16:57:27 -10:00

convert_vq_diffusion_to_diffusers.py

Update Ruff to latest Version (#10919)

2025-04-09 16:51:34 +05:30

convert_wan_to_diffusers.py

[Wan2.1-FLF2V] update conversion script (#11365)

2025-04-18 14:08:44 -10:00

convert_wuerstchen.py

Fix typos in docs and comments (#11416)

2025-04-30 20:30:53 -10:00

convert_zero123_to_diffusers.py

Remove dead code and fix f-string issue (#7720)

2024-05-08 13:15:28 -10:00

extract_lora_from_model.py

[chore] add a script to extract loras from full fine-tuned models (#10631)

2025-01-24 11:50:36 +05:30

generate_logits.py

Use model_info.id instead of model_info.modelId (#8912)

2024-07-20 20:01:21 +05:30