Pocket-Sized Multimodal AI for content understanding and generation across multilingual texts, images, and ๐ video, up to 5x faster than OpenAI CLIP and LLaVA ๐ผ๏ธ & ๐๏ธ
Terminal score: 0โ1000 raw, weighted across 4 dimensions. Public score: 0โ10 normalized (shown in the 30-day stars chart above).