Instructions to use OzzyGT/ERNIE_Image_Turbo_sdnq_dynamic_int4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use OzzyGT/ERNIE_Image_Turbo_sdnq_dynamic_int4 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("OzzyGT/ERNIE_Image_Turbo_sdnq_dynamic_int4", 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
- Local Apps Settings
- Draw Things
- DiffusionBee
Unrecognized keys and unexpected types
#1
by SamuelTallet - opened
Hello, thanks for this model quantization.
Inference works but I have these warnings in the background:
Unrecognized keys in `rope_parameters` for 'rope_type'='yarn': {'llama_4_scaling_beta'}
Expected types for text_encoder: (<class 'transformers.models.auto.modeling_auto.AutoModel'>,), got <class 'transformers.models.mistral3.modeling_mistral3.Mistral3Model'>.
Expected types for pe: (<class 'transformers.models.auto.modeling_auto.AutoModelForCausalLM'>, <class 'NoneType'>), got <class 'transformers.models.ministral3.modeling_ministral3.Ministral3ForCausalLM'>.
The first one is a transformers warning, I use 5.3 and I don't have it, but it's safe either way, the other two are from diffusers and they are also safe, it's just that the pipeline was made using automodels which is not how we usually do it, but the models load correctly.
You can safely ignore those warnings.
Okay, thank you π
SamuelTallet changed discussion status to closed