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

Apply suggestions from code review

Browse Source 
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
This commit is contained in:
Sayak Paul
2025-04-15 09:15:33 +05:30
committed by GitHub
parent 7b09d62e72
commit 9d36ebc227
3 changed files with 6 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
docs/source/en/quantization/torchao.md
View File

@@ -102,8 +102,7 @@ import torch
from diffusers import AutoModel, TorchAoConfig
quantization_config = TorchAoConfig("int8wo")
transformer = from diffusers import AutoModel, TorchAoConfig
.from_pretrained(
transformer = AutoModel.from_pretrained(
"black-forest-labs/Flux.1-Dev",
subfolder="transformer",
quantization_config=quantization_config,
@@ -151,11 +150,8 @@ with init_empty_weights():
transformer.load_state_dict(state_dict, strict=True, assign=True)
```
<Tip>
With Torch 2.6 or higher, you can directly do: `transformer = AutoModel.from_pretrained("/path/to/flux_uint4wo/")`.
</Tip>
> [!TIP]
> The [`AutoModel`] API is supported for PyTorch >= 2.6 as shown in the examples below.
## Resources

7
docs/source/en/quicktour.md
View File

@@ -163,11 +163,8 @@ Models are initiated with the [`~ModelMixin.from_pretrained`] method which also
>>> model = UNet2DModel.from_pretrained(repo_id, use_safetensors=True)
```
<Tip>
Use `AutoModel` from `from diffusers import AutoModel` in case you are unsure which model class to use.
</Tip>
> [!TIP]
> Use the [`AutoModel`] API to automatically select a model class if you're unsure of which one to use.
To access the model parameters, call `model.config`:

2
docs/source/en/using-diffusers/loading_adapters.md
View File

@@ -134,7 +134,7 @@ The [`~loaders.StableDiffusionLoraLoaderMixin.load_lora_weights`] method loads L
- the LoRA weights don't have separate identifiers for the UNet and text encoder
- the LoRA weights have separate identifiers for the UNet and text encoder
To directly load (and save) a LoRA adapter at the *model-level*, use [`~loaders.PeftAdapterMixin.load_lora_adapter`], which builds and prepares the necessary model configuration for the adapter. Like [`~loaders.StableDiffusionLoraLoaderMixin.load_lora_weights`], [`~loaders.PeftAdapterMixin.load_lora_adapter`] can load LoRAs for both the UNet and text encoder. For example, if you're loading a LoRA for the UNet, [`loaders.PeftAdapterMixin.load_lora_adapter`] ignores the keys for the text encoder.
To directly load (and save) a LoRA adapter at the *model-level*, use [`~loaders.PeftAdapterMixin.load_lora_adapter`], which builds and prepares the necessary model configuration for the adapter. Like [`~loaders.StableDiffusionLoraLoaderMixin.load_lora_weights`], [`~loaders.PeftAdapterMixin.load_lora_adapter`] can load LoRAs for both the UNet and text encoder. For example, if you're loading a LoRA for the UNet, [`~loaders.PeftAdapterMixin.load_lora_adapter`] ignores the keys for the text encoder.
Use the `weight_name` parameter to specify the specific weight file and the `prefix` parameter to filter for the appropriate state dicts (`"unet"` in this case) to load.