diffusers

diffusers copied to clipboard

torch.compile can't be used with groupoffloading on hunyuanvideo_frampack

3

### Describe the bug when i use block_level groupoffloading with use_stream=True on hunyuanvideo_framepack, i want to use torch compile to accelerate video generation. but it seems apply compile and groupoffloading...

zhangvia

OMI Format Compatibility

4

Opening up this thread to raise the conversation on compatibility with the Open Model Initiative standardized key format. We'd engaged w/ Marc Sun a while back on this topic, but...

hipsterusername

Error in init from pretrained for LTXConditionPipeline

2

### Describe the bug pipe = LTXConditionPipeline.from_pretrained("Lightricks/LTX-Video-0.9.7-dev", torch_dtype=torch.bfloat16) File "/opt/nas/p/conda/envs/diffuser0526/lib/python3.10/site-packages/huggingface_hub/utils/_validators.py", line 114, in _inner_fn return fn(*args, **kwargs) File "/opt/nas/p/conda/envs/diffuser0526/lib/python3.10/site-packages/diffusers/pipelines/pipeline_utils.py", line 986, in from_pretrained loaded_sub_model = load_sub_model( File "/opt/nas/p/conda/envs/diffuser0526/lib/python3.10/site-packages/diffusers/pipelines/pipeline_loading_utils.py", line 711,...

Vickeyhw

max_shard_size

1

### Describe the bug max_shard_size=1GB is set but the size of each safetensor is still greater than 1GB ### Reproduction pipe.save_pretrained(save_directory='FLUX.1-dev-Control-all', max_shard_size="1GB") ### Logs ```shell ``` ### System Info diffusers=0.30.2...

Ranking666

[WIP] [LoRA] support omi hidream lora.

5

# What does this PR do? Check https://github.com/huggingface/diffusers/issues/11653. This PR isn't at all ready. But opening up to discuss some doubts. Currently, this PR is only aimed at supporting the...

sayakpaul

Bump requests from 2.32.3 to 2.32.4 in /examples/server

1

Bumps [requests](https://github.com/psf/requests) from 2.32.3 to 2.32.4. Release notes Sourced from requests's releases. v2.32.4 2.32.4 (2025-06-10) Security CVE-2024-47081 Fixed an issue where a maliciously crafted URL and trusted environment will retrieve...

dependabot[bot]

The density_for_timestep_sampling and loss_weighting for SD3 Training！！！

17

Thanks to Rafie Walker's code we can try to train SD3 models with flow-matching! But some places don't seem to match what's in the paper. Rafie Walker's code is below:...

DidiD1

Attention Dispatcher

10

### Usage ```python # test.py import torch from diffusers import Lumina2Pipeline, attention_backend pipe = Lumina2Pipeline.from_pretrained("Alpha-VLLM/Lumina-Image-2.0", torch_dtype=torch.bfloat16) pipe.to("cuda") prompt = "A cat holding a sign that says 'Hello, World!' in a...

a-r-r-o-w

LoRA load issue

5

### Describe the bug FluxPipeline fails to load LoRA weights (https://civitai.com/models/876464): ### Reproduction ``` from diffusers import FluxPipeline from torch import bfloat16 prompt_from_repo = "Beautiful female woodelf in shiny iron...

D1-3105

Support Expert loss for HiDream

1

a-r-r-o-w