Instructions to use dg845/LTX-2.3-Diffusers with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use dg845/LTX-2.3-Diffusers with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("dg845/LTX-2.3-Diffusers", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
Getting the error: ValueError: Cannot load because transformer_blocks.0.audio_scale_shift_table expected shape torch.Size([6, 2048]), but got torch.Size([9, 2048]).
#2
by Phaserblast - opened
While trying to load this model using LTX2VideoTransformer3DModel, I'm getting the same error reported by folks when trying to load certain GGUF versions of this model:
ValueError: Cannot load because transformer_blocks.0.audio_scale_shift_table expected shape torch.Size([6, 2048]), but got torch.Size([9, 2048]).
I'm using transformers 5.5.4 and diffusers 0.37.1. Any idea of how to fix this?
Nevermind, I see your commits were made just after the 0.37.1 release. All good.
Phaserblast changed discussion status to closed