Back to feed
Hugging Face Blog·

Welcome PaliGemma 2 – New vision language models by Google

Signal
75
Hype
25
In three linesGoogle releases PaliGemma 2, a new family of open-source vision-language models. These models combine visual and textual capabilities for multimodal tasks, with versions optimized for different use cases.
Read source
Your take?
DeepMindVisionOpen sourceBenchmarks

Summary generated by Claude — human-verified