unum-cloud/uform

Pocket-Sized Multimodal AI for content understanding and generation across multilingual texts, images, and 🔜 video, up to 5x faster than OpenAI CLIP and LLaVA 🖼️ & 🖋️

16 best points · 1 comments · 4 Hacker News posts · 1,253 GitHub stars · View on GitHub

Hacker News discussions

• Show HN: UForm v2 – tiny CLIP-like embeddings in 21 languages and Graphcore API (16 points, 1 comments, by vov_or, 2023-08-18)
• Multimodal Chat in 1.5B Parameters (13 points, 0 comments, by ashvardanian, 2023-12-28)
• Show HN: UForm v2 Featuring Multimodal Matryoshka, Multimodal DPO, and ONNX (3 points, 0 comments, by ashvardanian, 2024-03-28)
• Multimodal Embeddings for JavaScript, Swift, and Python (2 points, 0 comments, by ashvardanian, 2024-04-25)