1
0
mirror of https://github.com/huggingface/diffusers.git synced 2026-10-27 21:33:02 +03:00

8 Commits

Author SHA1 Message Date
Sayak Paul
61a11f2191 [tests] refactor o and p series pipeline tests (#14646)
* refactor o and p series pipeline tests

* up
2026-09-01 08:34:33 +05:30
Sayak Paul
13a7bee487 [core] deprecate torch_dtype and prefer dtype following transformers. (#14205)
* deprecate torch_dtype and prefer dtype following transformers.

* address dhruv's comment on introducing a helper.
2026-07-21 07:32:08 +05:30
Sayak Paul
bc7aebc20a [core] deprecate dduf (#14141)
* deprecate dduf (re-apply after accidental revert on main).

* up
2026-07-16 13:24:12 +05:30
Steven Liu
23b0004330 torch.dtype alias (#14162)
* alias

* add tests

* feedback

* Fix failing test_dtype_alias pipeline tests

Cosmos pipelines keep the safety_checker in float32 and require it to be
passed explicitly on reload, mirroring their test_torch_dtype_dict
overrides. PRX skips dtype-dict/save-load tests because the custom
T5GemmaEncoder is not compatible with transformers v5; do the same for
test_dtype_alias.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-14 10:08:01 -07:00
David Bertoin
a487e2f131 Add PRXPixelPipeline: pixel-space PRX text-to-image pipeline (#13928)
Adds a pixel-space variant of PRX that denoises raw RGB directly
(no VAE), conditioned on a Qwen3-VL text encoder:

- PRXTransformer2DModel: new optional config args `bottleneck_size`
  (two-layer img_in projection for large patch dims) and
  `resolution_embeds` (PRXResolutionEmbedder conditions the timestep
  modulation on the latent resolution)
- PRXPipeline: support for subclass-tuned tokenizer max length, light
  text cleaning, x-prediction flow matching (x0 -> velocity conversion),
  and non-unit initial noise scale
- PRXPixelPipeline: thin subclass wiring the above together
  (vae optional/None, vae_scale_factor=1, 1024px default)
- conversion script support for the pixel checkpoint format
- registration in __init__ files + dummy objects, docs autodoc entry,
  fast pipeline tests

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-06-16 21:14:54 -07:00
Sayak Paul
5e94d62eb4 migrate to transformers v5 (#12976)
* switch to transformers main again./

* more

* up

* up

* fix group offloading.

* attributes

* up

* up

* tie embedding issue.

* fix t5 stuff for more.

* matrix configuration to see differences between 4.57.3 and main failures.

* change qwen expected slice because of how init is handled in v5.

* same stuff.

* up

* up

* Revert "up"

This reverts commit 515dd06db5.

* Revert "up"

This reverts commit 5274ffdd7f.

* up

* up

* fix with peft_format.

* just keep main for easier debugging.

* remove torchvision.

* empty

* up

* up with skyreelsv2 fixes.

* fix skyreels type annotation.

* up

* up

* fix variant loading issues.

* more fixes.

* fix dduf

* fix

* fix

* fix

* more fixes

* fixes

* up

* up

* fix dduf test

* up

* more

* update

* hopefully ,final?

* one last breath

* always install from main

* up

* audioldm tests

* up

* fix PRX tests.

* up

* kandinsky fixes

* qwen fixes.

* prx

* hidream
2026-02-24 10:53:56 +05:30
dg845
fe78a7b7c6 Fix ftfy import for PRX Pipeline (#13154)
* Guard ftfy import with is_ftfy_available

* Remove xfail for PRX pipeline tests as they appear to work on transformers>4.57.1

* make style and make quality
2026-02-18 20:44:33 -08:00
David Bertoin
dd07b19e27 Prx (#12525)
* rename photon to prx

* rename photon into prx

* Revert .gitignore to state before commit b7fb0fe9d6

* rename photon to prx

* rename photon into prx

* Revert .gitignore to state before commit b7fb0fe9d6

* make fix-copies
2025-10-21 17:09:22 -07:00