Instructions to use dataautogpt3/OpenDalleV1.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use dataautogpt3/OpenDalleV1.1 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("dataautogpt3/OpenDalleV1.1", dtype=torch.bfloat16, device_map="cuda") prompt = "black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
prompt
HI. very nice expressions!!! Thank u ..but... no model can do this simple prompt right:
"a couple on a date, sipping coffee. The man looks cheerful, but the woman seems unhappy"
they all fail ,and this one too. all sdxl models, and MidJourney also, fail. they are generating both man and woman happy or sad no matter how you rewrite the prompt.
ONLY dalle-3 can make it right
a challenge
We've reached the Dalle-3 level
Dalle is a pipeline. It doesn't just generate an image. It generates the image, makes the necessary corrections to the prompt and inpaints and then sends it to you. If you have the time, you can develop this.... Using an llm like mixtral to improve the prompt, clip-2 or openclip and something like ip2p for the corrections.