USearch adoption is snowballing — well beyond ClickHouse, DuckDB, YugaByte, TiDB & ScyllaDB open-source gurus and the vector search use-case to the contents of slide 41 of this 2023 lecture and further!. So many ways to exploit just a few thousand lines of C++ templates ⚛️

Although most of our open-source libraries focus on in-memory processing, truly large-scale processing typically begins in external memory. Here's a glimpse from 2022 into what has been in the works for quite some time and will hopefully be available to everyone 🔜

RT @ashvardanian: USearch everywhere 🎉.@Unum_Cloud 🤝 @Yugabyte

RT @ashvardanian: Higher recall + Python, JS, & Swift fixes🎉.Entirely community-driven - I did nothing 🤗.Upgrade to USearch v2.17.8 today 😉….

RT @ashvardanian: For everyone using @unum_cloud USearch. For <1B vectors per index you should aim for >100K insert & search queries/second….

RT @ashvardanian: One more DBMS project leveraging @unum_cloud USearch for Vector Search 🥳.

RT @ashvardanian: 20 million Python downloads across SimSIMD, StringZilla, and @unum_cloud USearch 🎉. If you’re chasing performance, upgrad….

RT @ashvardanian: Thanks to @PingCAP for trusting @unum_cloud USearch, opening a PR, and mentioning their project involvement!. Are there a….

Your pocket device is much more powerful than you know!. Together with our partners at @TheStageAI, we've accelerated the inference time of our UForm models by 5x, reaching only 0.5 ms for multimodal retrieval pipelines. locally, without exposing your privacy!. Details 👇

Rewinding to 2022. We shared our first major public talk on Linux kernel bypass & GPU-accelerated ACID-transactional storage — the result of 7 years of R&D. We hit 1M transactions per SSD and scaled linearly to 24 drives!. 🎥 Watch now:

Tech conferences come and go, but Unum is here for the long run!. We’re preserving some of the most impactful talks by re-uploading them to our official YouTube channel. Starting with "Vector Search and Databases at Scale" from 2023. 🎥 Watch now: 🧵.

RT @ashvardanian: The Next 31 Years of Developing Unum (@unum_cloud).

And the new inference code with binary and Matryoshka examples on our @GitHub repo - don't forget to 🌟 and spread the word 😉.

Check out the new collections on @huggingface and let us which languages we should prioritize for future multilingual releases 🤗.

Multimodal DPO, large & tiny multimodal matryoshka embeddings, and 1st party ONNX support for 10x lighter deployments 🥳. In partnership with @nebiusofficial, Unum is releasing a new set of pocket-sized multimodal models, already available on @huggingface 🤗

2023 was an excellent year for multimodal understanding and generation, but even the largest models still hallucinate. Committed to improving the quality of semantic search, we are looking to improve on that and share new models in 2024. Happy New Year!.