AI/diffusers
1
0
Fork 0
mirror of https://github.com/huggingface/diffusers.git synced 2026-01-29 07:22:12 +03:00
Code Activity

update

Browse Source
This commit is contained in:
DN6
2025-05-13 14:50:13 +05:30
parent 721375bb81
commit 2a19302e65
1 changed files with 14 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
docs/source/en/api/models/hidream_image_transformer.md
View File

@@ -26,24 +26,33 @@ transformer = HiDreamImageTransformer2DModel.from_pretrained("HiDream-ai/HiDream
GGUF checkpoints for the `HiDreamImageTransformer2DModel` can we be loaded using `~FromOriginalModelMixin.from_single_file`
```python
from diffusers import HiDreamImageTransformer2DModel
from diffusers import GGUFQuantizationConfig, HiDreamImageTransformer2DModel
ckpt_path = "https://huggingface.co/city96/HiDream-I1-Dev-gguf/blob/main/hidream-i1-dev-Q2_K.gguf"
transformer = HiDreamImageTransformer2DModel.from_single_file(ckpt_path, torch_dtype=torch.bfloat16)
transformer = HiDreamImageTransformer2DModel.from_single_file(
ckpt_path,
quantization_config=GGUFQuantizationConfig(compute_dtype=torch.bfloat16),
torch_dtype=torch.bfloat16
)
```
If you are trying to use a GGUF checkpoint for the `HiDream-ai/HiDream-E1-Full` model, you will have to pass in a `config` argument to properly configure the model. This is because the HiDream I1 and E1 models share the same state dict keys, so it is currently not possible to automatically infer the model type from the checkpoint itself.
```python
from diffusers import HiDreamImageTransformer2DModel
from diffusers import GGUFQuantizationConfig, HiDreamImageTransformer2DModel
ckpt_path = "https://huggingface.co/ND911/HiDream_e1_full_bf16-ggufs/blob/main/hidream_e1_full_bf16-Q2_K.gguf"
transformer = HiDreamImageTransformer2DModel.from_single_file(ckpt_path, config="HiDream-ai/HiDream-E1-Full", subfolder="transformer", torch_dtype=torch.bfloat16)
transformer = HiDreamImageTransformer2DModel.from_single_file(
ckpt_path,
quantization_config=GGUFQuantizationConfig(compute_dtype=torch.bfloat16),
config="HiDream-ai/HiDream-E1-Full",
subfolder="transformer",
torch_dtype=torch.bfloat16
)
```
## HiDreamImageTransformer2DModel
[[autodoc]] HiDreamImageTransformer2DModel