JinjieNi/MixEval-X

The official github repo for MixEval-X, the first any-to-any, real-world benchmark.

19
/ 100
Experimental

This project helps AI researchers and developers accurately compare the real-world performance of different large AI models, especially those capable of handling various types of input and output. It takes the model's responses to diverse prompts (like images, videos, audio, or text) and outputs a comprehensive score that reflects how well the model performs on real-world tasks. The primary users are researchers and engineers developing or evaluating large multimodal models.

No commits in the last 6 months.

Use this if you need a standardized, comprehensive, and efficient way to benchmark the real-world performance of your multimodal AI models against a diverse set of tasks and modalities.

Not ideal if you are looking for a tool to train models or if your primary focus is on single-modality evaluations without a need for real-world, multimodal task distributions.

AI model evaluation Multimodal AI Benchmarking Generative AI Model comparison
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 8 / 25
Community 5 / 25

How are scores calculated?

Stars

16

Forks

1

Language

Python

License

Last pushed

Feb 15, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/JinjieNi/MixEval-X"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.