With over 23 million downloads and 956 likes, CLIP is a community favorite. It achieves strong zero shot performance on ImageNet and other benchmarks, rivaling fully supervised models. It's open source, endpoint compatible, and ready to deploy today.
Imagine a model that can look at any image and tell you what's in it, without needing specific training. That's CLIP. It's a game changer for zero shot image classification, understanding images and text together. No more endless labeling. Just pure AI magic.
Over 22 million downloads and 1,300 likes say it all. Apache 2.0 license means free to use, modify, and deploy. It's a top-tier choice for lightweight conversational AI, proving size isn't everything when it comes to performance. #OpenSource#AI
Ever wondered how a tiny 0.6B parameter model can handle complex conversations? Meet Qwen3-0.6B, a compact text-generation powerhouse from the Qwen family. It's built for speed and smarts, perfect for developers who want big results without the heavy compute. #AI#LLM
Outperforms mBERT on XNLI and other benchmarks, with 21.9M downloads and 843 likes. It's a top multilingual model, supporting 100 languages with state-of-the-art accuracy. Try it for your next global AI project!
Ever wanted one AI that speaks 100 languages? Meet XLM-RoBERTa, a multilingual beast trained on 2.5TB of text across 100 languages. It fills in missing words like a polyglot, perfect for global NLP tasks. #AI#NLP
With over 20 million downloads and 103 likes, CLAP is a community favorite. It achieves state of the art on zero shot audio classification benchmarks like AudioSet and ESC 50. Fast, accurate, and open source.
This model uses a HTSAT audio encoder fused with a text encoder, trained on 1.2 million audio text pairs. It's built with transformers and safetensors for efficient inference. Zero shot means no fine tuning needed for new sounds.
Build apps that search audio by text, classify sounds in real time, or retrieve clips from a library. Perfect for content moderation, music discovery, and accessibility tools. Just describe what you need and CLAP finds it.
Imagine an AI that can 'hear' and understand audio like we do. That's CLAP! It's a zero shot audio classifier that can recognize sounds, music, and speech without being trained on specific examples. Game changer for audio AI.
79 Followers 20 FollowingCo-founder @savink_app 💻 | Accidentally running a company | Building cool AI + open source stuff. @cosmicstack | Hiring - ✉️ DM. Research @mercury__agent
84K Followers 0 FollowingThe official home of Google's Gemma. Lightweight, state-of-the-art open models by Google DeepMind, built on Gemini tech. What will you build? 🚀💻
14K Followers 256 FollowingAn open-source AI creative engine for real production. Native 4K fidelity, built for enterprise, developers, and studios. By @lightricks
7K Followers 23 Followinghttps://t.co/zI71a4QB1W
https://t.co/QFKNuHms1N
[email protected]
Donation: Support our work on Ko-fi (https://t.co/gAtHKPSCHH)!
76K Followers 1K FollowingCo-founder and CTO at @huggingface 🤗. ML/AI for everyone, building products to propel communities fwd. @Stanford + @Polytechnique
72K Followers 139 FollowingHave questions, or building something cool with Cloudflare's Developer products? We're here to help. For help with your account please try @CloudflareHelp
82K Followers 145 FollowingAnnouncing new open source releases, exploring projects, sharing how we approach FOSS, and supporting communities around the world.
324K Followers 64 FollowingWe're sharing/showcasing best of @github projects/repos. Follow to stay in loop. Promoting Open-Source Contributions. UNOFFICIAL, but followed by github