j-min/DallEval
DALL-Eval: Probing the Reasoning Skills and Social Biases of Text-to-Image Generation Models (ICCV 2023)
This tool helps researchers and evaluators understand how well text-to-image AI models interpret and generate visuals from text prompts. You input a text prompt and an image generated by an AI, and it outputs a detailed assessment of the image's quality, visual reasoning skills, and potential social biases (gender and skin tone). This is for AI researchers, ethicists, or product managers working with generative AI.
143 stars. No commits in the last 6 months.
Use this if you need to systematically evaluate the performance and fairness of text-to-image generative AI models.
Not ideal if you are looking to fine-tune or train a text-to-image model, or if you only need a subjective evaluation of image aesthetics.
Stars
143
Forks
6
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Jun 10, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/j-min/DallEval"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ai-forever/ru-dalle
Generate images from texts. In Russian
KR-HappyFace/KoDALLE
🇰🇷 Text to Image in Korean
olaviinha/NeuralTextToImage
Colabs for text prompt steered image generators
kevinhermawan/Canvas
DALL·E playground for the Mac
simonsanvil/DALL-E-Explained
Description and applications of OpenAI's paper about DALL-E (2021) and implementation of other...