Tagged: multimodal-ai

22 articles

Latest

Google's Gemini Omni 1.1 Flash reads 10 seconds of context to fix AI video's biggest flaw

Inside CLIP and friends: How vision-language models actually learn

Alibaba's Wan3.0 doubles AI video length to 30 seconds and accepts PDFs as prompts

Hugging Face releases IDEFICS, an 80B open-source rival to Flamingo

Adobe Firefly Now Generates Music, Speech, and SFX—All Cleared for Commercial Use

No AI model can truly see: best score on new perception test is 59.7%

Meta’s 30B Muse Glimmer launches open-source: A 2B vision encoder meets hybrid attention

PRISM2 reads 2.3M pathology slides and talks back, beating clinical-grade cancer detectors

MoonshotAI’s PerceptionBench Exposes Where GPT-4o-Mini Still Can’t See Straight

MiniMax H3 folds 6 video models into one and ships 2K for $1.95 a clip

Thinking Machines drops Inkling: a 975B-param open model that sees, hears, and reads

TimeScope Exposes a Harsh Truth: Most AI Models Can't Really Understand Hour-Long Videos

Hugging Face drops nanoVLM: a 135M-param vision model you can train from scratch in pure PyTorch

AI’s new Swiss Army knives: VLMs are now reasoning, talking, and running on phones

Salamandra 7B gains vision via SigLIP, trained on 6.1M multilingual samples

Gemma 3's 27B model punches into top 10 of Chatbot Arena, beating Gemini 1.5 Pro

Hugging Face and IISc aim to fix AI's India problem with 150K hours of speech

Google's SigLIP 2 adds a decoder and self-distillation to build vision encoders that actually know where things are

How 1.9M YouTube videos became FineVideo's 44K highly-annotated training gem

Prezi leans on Hugging Face experts to sharpen its AI presentation pipeline

Hugging Face's Idefics2 shrinks to 8B parameters but now beats much larger vision models

Qwen2-VL and 7 others: The open VLM field just got crowded