🦀 @huggingface just released Text Embeddings Inference (TEI) v1.7.3 with efficient and performant support for Qwen3-Embedding models in CPU, MPS, CUDA, and Intel HPUs!. Qwen3-Embedding (0.6B, 4B, 8B) open-weights were recently released by @Alibaba_Qwen, and are killing it in the

RT @_lewtun: Really excited to share SmolLM3: a strong, smol reasoner!. > SoTA 3B model.> dual mode reasoning (think/no_think).> long conte….

Out of curiosity, has anyone ever used this feature from @github?. I constantly see it on the landing page, but never used it or had the need, so willing to know if there are any specific use-cases or scenarios where you find it useful (?)

@huggingface @Alibaba_Qwen @xenovacom 🙏 This release wouldn't have been possible without the help from our amazing community contributors!. - (first-time contributor).- (active contributor).- (first-time contributor).-

@huggingface @Alibaba_Qwen 💻 Alternatively, you can also leverage the ONNX runtime with Transformers JS (@xenovacom) in the browser!.

@huggingface @Alibaba_Qwen 🤏🏻 Qwen3-Embedding-0.6B **only** has 600M parameters, meaning it can run on-device and even on some scenarios with reduced compute, and with the latest TEI release you can leverage the ONNX backend too on CPU!

@huggingface @Alibaba_Qwen 😱 Qwen3-Embedding models are taking on the MTEB Leaderboard, closely competing with Gemini Embedding i.e., gemini-embedding-001; whilst being open, with a permissive Apache 2.0 license, and ranging sizes 0.6B, 4B and 6B.

@huggingface @Alibaba_Qwen 🤗 Qwen3-Embedding release happened in the @huggingface Hub, and you can check all the variants (including GGUF quants) in their Hub Collection!.

RT @jeffboudier: Welcome @Azure to @huggingface docs! 🤗. Find guides to deploy one of the 10,000+ open models available for 1-click deploy….

Then, weeks or months after having stopped taking notes, I miss it and come back to it (but on either a different application or starting from blank i.e., not continuing where I left off). Any advice?.

I really enjoy taking notes, having everything written down, where it should, being able to read it later, etc., but I struggle on handling notes when those get complex, or just after N time I stop taking notes for no reason. .

Is there any book, post, advice, or whatever that helps you be better and, specifically, more consistent, on note taking?.

@huggingface @Azure We can’t wait to see what you create with open models & open-source on Azure!.

@huggingface @Azure 🌐 Tips for integrating models into Azure ML workflows & building generative AI apps. 🤝 We work closely with Azure teams to bring high-quality open models into the Azure AI catalog—ready for your next GenAI project. 🔒 All models on Azure AI/ML use Safetensors weights &.

@huggingface @Azure ⚡ Step-by-step guides for deploying open models on Azure AI Foundry & Azure ML—no custom code needed. 🤖 Examples for building agentic AI: combine LLMs with tools & APIs. 🛠️ Tutorials for using open-source inference (TGI, TEI, etc) with OpenAI-compatible API routes.

🔵 Almost 3 years ago, @huggingface & @Azure teamed up to make open-source & open-model AI accessible to everyone on Microsoft Azure!. Today, we’re excited to announce brand new docs & examples for building with the latest open models on Azure! 👇

I've been saying this over and over already, but Unsloth's content is *chef kiss* 😙🤌.

RT @UnslothAI: We made a complete Guide on Reinforcement Learning for LLMs!. Learn about:.• RL's goal & why it's key to building intelligen….

RT @GoogleColab: 🤗The future of ML is accessible & collaborative 🤝.We’ve partnered with Hugging Face to add “Open in Colab” support for all….

🤗 Today marks 1 year since I joined @huggingface, and it’s been an incredible journey so far — full of learning, growth, and meeting amazing people!. Over the past year, I’ve had the chance to work on bridging the gap between Hugging Face and some of our partners, such as