TBNilles/flux2
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update 4-bit text-encoder and 4-bit DiT on diffusers script for initial CPU loading (#12)

Browse Source 
* Update diffusers script for initial CPU loading

* remove transformer
This commit is contained in:
apolinário
2025-11-28 20:28:31 +01:00
committed by GitHub
parent 6bb103559d
commit c7a09571ba
2 changed files with 11 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@@ -88,7 +88,7 @@ The below example should run on a RTX 4090. For more examples check the [diffuse
```python
import torch
from diffusers import Flux2Pipeline, Flux2Transformer2DModel
from diffusers import Flux2Pipeline
from diffusers.utils import load_image
from huggingface_hub import get_token
import requests
@@ -112,7 +112,7 @@ def remote_text_encoder(prompts):
return prompt_embeds.to(device)
pipe = Flux2Pipeline.from_pretrained(
repo_id, transformer=transformer, text_encoder=None, torch_dtype=torch_dtype
repo_id, text_encoder=None, torch_dtype=torch_dtype
).to(device)
prompt = "Realistic macro photograph of a hermit crab using a soda can as its shell, partially emerging from the can, captured with sharp detail and natural colors, on a sunlit beach with soft shadows and a shallow depth of field, with blurred ocean waves in the background. The can has the text `BFL Diffusers` on it and it has a color gradient that start with #FF5733 at the top and transitions to #33FF57 at the bottom."