Instructions to use Abiray/ERNIE-Image-Turbo-FP8-NVFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Abiray/ERNIE-Image-Turbo-FP8-NVFP4 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Abiray/ERNIE-Image-Turbo-FP8-NVFP4", 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
Text Encoder nvfp4 ?
#2
by ireMaN - opened
i know prompt enhancer file works as t_e, but is there a real quantized text encoder ?
ERNIE-Image-Turbo actually uses Ministral-3-3B as its text encoder, but I don't think anyone has created a GGUF version for it yet. You can use this Hugging Face repo https://huggingface.co/InsecureErasure/ministral-3-3b-TE-only-ComfyUI if you want. A GGUF file would have to be created manually, and honestly, I doubt it would perform well. If I find the time, I'll think about making a GGUF for it myself.