I went together with Bo to SIGIR this year, we wrote a blog post with our highlights and summaries of AI and neural papers that we found interesting at the conference jina.ai/news/what-we-l…

Two weeks ago, we released jina-embeddings-v4-GGUF with dynamic quantizations. During our experiments, we found interesting things while converting and running GGUF embeddings. Since most of the llama.cpp community focuses on LLMs, we thought it'd be valuable to share this from

Got a Mac with an M-chip? You can now train Gemma3 270m locally as a multilingual embedding or reranker model using our mlx-retrieval project. It lets you train Gemma3 270m locally at 4000 tokens/s on M3 Ultra - that's actually usable speed. We've implemented some standard

We are at Qdrant 's Vector Space Day 🚀 in Berlin on Sep 26. We'll talk about "Vision-Language Models: A New Architecture for Multi-Modal Embedding Models" and also share some insights and learnings we gained while training jina-embeddings-v4. 🎫 lu.ma/p7w9uqtz

Today we're releasing jina-code-embeddings, a new suite of code embedding models in two sizes—0.5B and 1.5B parameters—along with 1~4bit GGUF quantizations for both. Built on latest code generation LLMs, these models achieve SOTA retrieval performance despite their compact size.

mmBERT: Massively Multilingual BERT Trained on 3T+ tokens across 1,833 languages, mmBERT surpasses XLM-R on standard NLU and retrieval benchmarks and is competitive with English-only encoders; in throughput tests it runs 2–4× faster than prior multilingual encoders under

V4 is multimodal embeddings, but V4-GGUF wasn't—until now. We've finally cracked how to generate multimodal embeddings using llama.cpp & GGUF. We fixed two main issues. First, in the language model part, we corrected the attention mask in the transformer block so it properly

Last but not late: jina-reranker-v3 is here! A new 0.6B-parameter listwise reranker that puts query and all candidate documents in one context window and SOTA on BEIR. We call this new query-document interaction "last but not late" - It's "last" because <|doc_emb|> is placed as

Heard you like GGUFs and MLX. Our newly released listwise reranker, jina-reranker-v3, is now available in dynamic quantized GGUFs and MLX. Check out our🤗 collection for the weights and arxiv report: huggingface.co/collections/ji…

We’re excited to announce that we have joined forces with Jina AI, a leader in frontier models for multimodal and multilingual search. This acquisition deepens Elastic’s capabilities in retrieval, embeddings, and context engineering to power agentic AI: go.es.io/48QeYCM

kudo to NEXA AI 👍, your favorite jina-reranker-v2 by Jina AI can run on Qualcomm NPU 👏! This unlocks a lot of imaginations 🔥.

Does synthetic data always help text-embedder models? Not quite. The gains are sparse and come with trade-offs. We open-source data + code to make research on synthetic data for embeddings more rigorous. 1/

The MTEB team has just released MTEB v2, an upgrade to their evaluation suite for embedding models! Their blogpost covers all changes, including easier evaluation, multimodal support, rerankers, new interfaces, documentation, dataset statistics, a migration guide, etc. 🧵

In 2 weeks, we're presenting at #EMNLP2025 and hosting a BoF on Embeddings, Rerankers, Small LMs for Better Search, again! Come check out our research on training data for multi-hop reasoning, multimodal embeddings, and where retrieval models are headed in 2025/26. Say hi to our

I'm organizing BoF EMNLP 2025 again, always a fun experience - see u in Suzhou, China!

📢 ViDoRe V3, our new multimodal retrieval benchmark for enterprise use cases, is finally here! It focuses on real-world applied RAG scenarios using high-quality human-verified data. huggingface.co/blog/QuentinJG… 🧵(1/N)

Jina-V4 is a legend But our API sucks a bit🥲

My colleague Saba Sturua presents jina-embddings-v4 and the JinaVDR benchmark for visual document retrieval EMNLP 2025 and the MRL workshop