linzhiqiu/t2v_metrics
Evaluating text-to-image/video/3D models with VQAScore
This tool helps researchers and engineers quickly and automatically assess how well their text-to-image, video, or 3D models generate visuals that accurately match a given text description. You input a text prompt and the visual output from a generative model, and it provides a score indicating the alignment between the two. It's designed for machine learning researchers and AI engineers developing or fine-tuning generative visual AI models.
381 stars. No commits in the last 6 months.
Use this if you need an automated, robust, and efficient way to evaluate the quality and compositional accuracy of your text-to-visual generative AI models, especially when human evaluation or proprietary models are too slow or costly.
Not ideal if you are looking for a tool to generate images or videos, or if your primary goal is to evaluate the aesthetic quality or realism of generated content rather than its alignment to a text prompt.
Stars
381
Forks
34
Language
Jupyter Notebook
License
Apache-2.0
Category
Last pushed
Sep 22, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/generative-ai/linzhiqiu/t2v_metrics"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
open-mmlab/mmagic
OpenMMLab Multimodal Advanced, Generative, and Intelligent Creation Toolbox. Unlock the magic 🪄:...
jdh-algo/JoyVASA
Diffusion-based Portrait and Animal Animation
haidog-yaqub/EzAudio
High-quality Text-to-Audio Generation with Efficient Diffusion Transformer
404-Repo/404-gen-blender-add-on
Blender add-on for 404-GEN 3D generator running on Bittensor
TIGER-AI-Lab/AnyV2V
Code and data for "AnyV2V: A Tuning-Free Framework For Any Video-to-Video Editing Tasks" [TMLR 2024]