LLaVA-Mini: Efficient Image and Video Large Multimodal Models with One Vision Token · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2501.03895/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTLLaVA-Mini: Efficient Image and Video Large Multimodal Models with One Vision Token — mentioned by Vilyhttps://news.ycombinator.com/item?id=42679455urn:hackerpapers:mention:2501.03895:42679455Mon, 13 Jan 2025 02:35:04 GMT<p>LLaVA-Miniis an efficient LMM for image/video understanding using 1 vision token, offering: (1)fast response (40ms per image), (2)less VRAM usage (support 3-hour video understanding on 24GB GPU). Paper: https://arxiv.org/abs/2501.03895 Code &amp; Demo: https://github.com/ictnlp/LLaVA-Mini</p><p><a href="https://hackerpaper.kaipfister.de/p/2501.03895/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CV