New benchmarks in vision-language models for real-world use: Google Research

[@RustemShaimagambetov]

NIce video, but how about qwen-vl-max, qwen-vl-plus. Did you test it?

[@yonatanbitton6386]

Thank you very much for the response!
No, these models were released after we ran our experiements, but in both projects the auto-eval code is available, and anyone can test and submit their results :)