🐛 I've just released Sentence Transformers v5.1.1! It's a small patch release that makes the project more explicit with incorrect arguments and introduces some fixes for multi-GPU processing, evaluators, and hard negatives mining. Details in 🧵

ModernBERT goes MULTILINGUAL! One of the most requested models I've seen, @jhuclsp.bsky.social has trained state-of-the-art massively multilingual encoders using the ModernBERT architecture: mmBERT. Stronger than an existing models at their sizes, while also much faster! Details in 🧵

Google just launched EmbeddingGemma: an efficient, multilingual 308M embedding model that's ready for semantic search & more on just about any hardware, CPU included. Details in 🧵:

One of the most underrated players in AI models, IBM, released 2 new extremely efficient embedding models: granite-embedding-english-r2 & granite-embedding-small-english-r2, commercially viable. Details in 🧵:

😎 I just published Sentence Transformers v5.1.0, and it's a big one. 2x-3x speedups of SparseEncoder models via ONNX and/or OpenVINO backends, easier distillation data preparation with hard negatives mining, and more! See 🧵for the deets:

OpenAI is back with open source releases on @hf.co. This is the biggest release I've seen in a long time! See more in huggingface.co/openai

I've just updated SetFit to v1.1.3, bringing compatibility with the recent datasets v4.0+ and Sentence Transformers v5.0+. You'll again be able to train tiny classifiers using very little training data! 🧵

Some of the ModernBERT team is back with new encoder models: Ettin, ranging from tiny to small: 17M, 32M, 68M, 150M, 400M & 1B parameters. They also trained decoder models & checked if decoders could classify & if encoders could generate. Details in 🧵:

‼️Sentence Transformers v5.0 is out! The biggest update yet introduces Sparse Embedding models, encode methods improvements, Router module for asymmetric models & much more. Sparse + Dense = 🔥 hybrid search performance! Details in 🧵

ColBERT (a.k.a. multi-vector, late-interaction) models are extremely strong search models, often outperforming dense embedding models. And @lightonai.bsky.social just released a new state-of-the-art one: GTE-ModernColBERT-v1! Details in 🧵

I just released Sentence Transformers v4.1; featuring ONNX and OpenVINO backends for rerankers offering 2-3x speedups and improved hard negatives mining which helps prepare stronger training datasets. Details in 🧵

I just published the SetFit v1.1.2 release, allowing you to finetune efficient classification models with very little training data. The new patch introduces compatibility with the latest transformers and sentence-transformers versions.

I just released Sentence Transformers v4.0.2. This patch release brings a slight improvement for maximum sequence lengths, typing issues, and distributed training device placement. Details in 🧵

I've just ported the excellent monoELECTRA-{base, large} reranker models from @fschlatt.bsky.social & the research network Webis Group to Sentence Transformers! These models were introduced in the Rank-DistiLLM paper, and distilled from LLMs like RankZephyr and RankGPT4. Details in 🧵

Ever thought that finetuning rerankers for your search system wasn't worth it? Think again. I finetuned ModernBERT on my dataset, TriviaQA, just on my normal PC in a few hours. It outperforms all open models on the market on my eval set. Finetuning is so worth it. Blog in 🧵

‼️Sentence Transformers v4.0 is out! You can now train and finetune reranker (aka cross-encoder) models with multi-GPU training, bf16 support, loss logging, callbacks & much more. I also prove that finetuning on your domain helps much more than you might think. Details in 🧵

Let's goo, Mixedbread is back with more open weight Apache 2.0 rerankers. Their new mxbai-rerank-...-v2 line is quite novel, using GRPO (akin to DeepSeek-R1), contrastive learning, and ranking objectives. Details in 🧵

I just tried to find a paper from a few months ago for some docs that I'm writing. I tried on Google with like 5 separate queries, but just couldn't find it. Tried it again via hf.co/papers, same prompt, and the paper I wanted is literally the first result. 🩷🤗🩷

An assembly of 18 European companies, labs, and universities have banded together to launch 🇪🇺 EuroBERT! It's a state-of-the-art multilingual encoder for 15 European languages, designed to be finetuned for retrieval, classification, etc. Details in 🧵

Already 1k likes for QwQ-32B on @hf.co in <24 hours! Well deserved, too.

The fine folks at Qwen released QwQ-32B 30 minutes ago, rivaling DeepSeek-R1 671B on various benchmarks, and outperforming OpenAI o1-mini on several as well! Vibe checks still in the works. Details & Links in 🧵

We've just released MMTEB, our multilingual upgrade to the MTEB Embedding Benchmark! It's a huge collaboration between 56 universities, labs, and organizations, resulting in a massive benchmark of 1000+ languages, 500+ tasks, and a dozen+ domains. Details in 🧵

The folks at Nomic just released nomic-embed-text-v2-moe, a MoE (Mixture of Experts) embedding model with state-of-the-art retrieval performance across ~100 languages. I've never seen a MoE embedding model until now, so very nice to see these! More in 🧵

Today, The Minish Lab released 2 more Static Embedding models; potion-base-32M & potion-retrieval-32M stronger performance than before, while still easily processing e.g. 50k sentences per second. The text embeddings can be used for retrieval, classification, clustering, etc. 🧵

Sentence Transformers v3.4.1 is out! A nice and small release bringing you instant access to some nice and small models: Model2Vec compatibility! These types of models are extremely fast (e.g. 400x faster than small transformer-based models), while only being a tad worse. 🧵

I just released Sentence Transformers v3.4.0, featuring a memory leak fix (memory not being cleared upon model & trainer deletion), compatibility between the powerful Cached... losses and the Matryoshka loss modifier, and a bunch of fixes & small features. Details in 🧵

🏎️ Today I'm introducing a method to train static embedding models that run 100x to 400x faster on CPU than common embedding models, while retaining 85%+ of the quality! Including 2 models with training scripts, datasets, metrics, evals, ideation, all public. Details in 🧵

ModernBERT-embed-base is awesome because it allows to use ModernBERT-base for various tasks out-of-the-box But the large variant of ModernBERT is also awesome... So today, @lightonai.bsky.social is releasing ModernBERT-embed-large, the larger and more capable iteration of ModernBERT-embed!

The newest extremely strong embedding model based on ModernBERT-base is out: `cde-small-v2`. Both faster and stronger than its predecessor, this one tops the MTEB leaderboard for its tiny size! Details in 🧵

That didn't take long! Nomic AI has finetuned the new ModernBERT-base encoder model into a strong embedding model for search, classification, clustering and more! Details in 🧵

BERT is BACK! I joined a collaboration with AnswerAI and LightOn to bring you the next iteration of BERT. Introducing ModernBERT: 16x larger sequence length, better downstream performance (classification, retrieval), the fastest & most memory efficient encoder on the market. 🧵

👀something's coming...

The Lichess database of games, puzzles, and engine evaluations is now on @hf.co - https://huggingface.co/Lichess. Billions of chess data points to download, query, and stream and we're excited to see what you'll build with it! ♟️ 🤗

❄️ Snowflake just released 2 very strong yet small embedding models for English and Multilingual retrieval. Large model: huggingface.co/Snowflake/sn... Medium model: huggingface.co/Snowflake/sn... Details in 🧵

Check out the most liked and downloaded models from 2022 to 2024 on Hugging Face: huggingface.co/spaces/huggi... Tomorrow we'll have another visualization!

If it takes me a minute to respond to your internship application, this is probably the reason why! Having said that, you can still apply to implement sparse models with me: - Remote, EMEA timezones: apply.workable.com/huggingface/... - Remote, US timezones: apply.workable.com/huggingface/...

If you're interested in embedding models for retrieval (search), clustering, classification, paraphrase mining, etc., then there's now 10,000 fully free and open source options on @hf.co via Sentence Transformers. Check out the most popular ones here: huggingface.co/models?libra...

I'm looking for an intern to introduce Sparse Embedding models to Sentence Transformers! If you're passionate about open source, interested in helping practitioners use your tools, and enjoy embedders/retrievers/rerankers, then I'd love to hear from you! Links with details and to apply in 🧵

Small yet mighty! 💫 We are releasing SmolVLM: a new 2B small vision language made for on-device use, fine-tunable on consumer GPU, immensely memory efficient 🤠 We release three checkpoints under Apache 2.0: SmolVLM-Instruct, SmolVLM-Synthetic and SmolVLM-Base huggingface.co/collections/...

First dataset for the new @huggingface.bsky.social @bsky.app community organisation: one-million-bluesky-posts 🦋 📊 1M public posts from Bluesky's firehose API 🔍 Includes text, metadata, and language predictions 🔬 Perfect to experiment with using ML for Bluesky 🤗 huggingface.co/datasets/blu...

New research by GovTech Singapore trains both bi-encoder and cross-encoder models for "Off-Topic Prompt Detection". Details: 1️⃣ Set a problem space, e.g. define acceptable/unacceptable behaviours 2️⃣ Generate prompts using an LLM 3️⃣ Train lightweight classifiers on this dataset Links in 🧵

Almost 10k public Sentence Transformer models. Customized, finetuned embedding models for the win! 🎉 Find all pretrained models: huggingface.co/models?libra... Learn how to finetune them: sbert.net/docs/sentenc...

⚠️ Work in Progress! ⚠️ I'm experimenting with IPEX (intel-extension-for-pytorch) as an inference backend for Sentence Transformers embedding models on CPU. The first tests aren't stellar - both ONNX and OpenVINO seem to perform better. I wonder if perhaps I'm not using the optimizations correctly.

✨ Jina AI just released Jina-CLIP-v2: A multimodal (images and texts) & multilingual embedding model. Details in 🧵 Model: huggingface.co/jinaai/jina-... 📈 Jina-CLIP-v2 outperforms Jina-CLIP-v1 (by 3% on text-image and text-text tasks) 🧵

Don't forget to set your Bluesky account in your @huggingface.bsky.social profile! Instructions in 🧵

I created a collection with good models for dataset curation - NSFW classifiers - PII classifiers - blazing fast embeddings by model2vec - quality classifier - educational value classifier - domain classifier Collection: huggingface.co/collections/...

Working on finetuning the nvidia/NV-Embed-v2 embedding model with PEFT for the new Sentence Transformers 🤝 PEFT Documentation. Excited to see if this works well - I don't usually finetune such large embedding models.

500 followers already - let's keep it up so I don't have to go to the other site anymore

I've trained 2 interesting embedding models over the 2 last months - I've just had to wait with the release blogpost for Sentence Transformers v3.3.0 for the NanoBEIREvaluator. Now that the evaluator is out, I can start preparing for release!

We need more open-source models, datasets & apps in AI healthcare. It makes it not only more likely to benefit everyone but also much safer, controllable and respects privacy! Great initiative so that everyone can build their own open-source AI health tracker for $12: huggingface.co/blog/cyrilza...