tusKANNy is a research team building state-of-the-art algorithms and open-source tools for ANN search across dense, sparse, and multi-vector neural representations.
High-performance graph-based indexes and quantization techniques for efficient search over dense embeddings.
Efficient inverted-index and graph structures for learned sparse embeddings, reaching near-exact accuracy at microsecond latency.
State-of-the-art retrieval pipelines that handle multi-vector representations without sacrificing speed.
Latest
Writing
Our two entries to Task 3 of the SISAP 2026 Indexing Challenge, a kANNolo graph and a Seismic inverted index built on the same foundation, finished first and second.
RerankIndex implements two-stage retrieval as a single building block. kANNolo 0.12 ships dense and sparse two-stage indexes built on it, up to 4.3× faster than plain HNSW.
Open Source
Fast and scalable late-interaction multi-vector retrieval
TACHIOM is a fast and scalable data structure for late-interaction multi-vector retrieval, written in Rust with Python bindings. It allows to cluster hundreds of millions of vectors in few minutes and to retrieve from large multivector collections in under 10 ms.
Fast approximate retrieval over learned sparse embeddings
A superluminal search engine for learned sparse representations, written in Rust with Python bindings. Seismic indexes sparse vector collections and retrieves results in microseconds while maintaining near-exact accuracy.
Fast ANN library for dense, sparse, and multi-vector embeddings
A flexible Rust library combining state-of-the-art indexing techniques for both dense, sparse, and multivector embeddings. Designed to make prototyping new ANN algorithms fast and ergonomic.
Unified embedding storage and compression backbone
A Rust library for storing, accessing, and compressing dense, sparse, and multi-vector embedding datasets. Provides a unified dataset/encoder interface shared across tusKANNy's indexing and search crates. Includes an exhaustive search API and a CLI tool for ground-truth computation.
Elementary yet very fast exact index for learned sparse retrieval
An exact inverted index for learned sparse representations, written in Rust. It scores every posting with no pruning or approximation, using a cache-friendly term-at-a-time strategy over doc-id blocks. Ideal for ground-truth computation: on MS MARCO it is 37× faster than a multithreaded brute-force scan, index build included.
Research
Silvio Martinico, Franco Maria Nardini, Cosimo Rulli, Rossano Venturini
Describes how the kANNolo library and the Seismic index share a common foundation, showing how to combine them for efficient approximate retrieval over learned sparse representations.
Matteo Attimonelli, Alessandro De Bellis, Franco Maria Nardini, Claudio Pomo, Cosimo Rulli, Rossano Venturini, Tommaso Di Noia
Sparsifies pre-trained universal multimodal embeddings after training, enabling efficient retrieval with sparse indexes without retraining the embedding model.
Franco Maria Nardini, Luca Rizzo, Cosimo Rulli, Rossano Venturini
Presents compact index layouts for learned sparse retrieval that reduce memory footprint while preserving query efficiency and accuracy.
Francesco Gargiulo, Rossano Venturini
Introduces OnPair, a dictionary-based compression algorithm for short strings that builds its dictionary in a single cache-friendly pass and compresses each string independently, matching BPE compression ratios with much faster training and fine-grained random access.
Community
A curated list of papers, libraries, benchmarks, and tools for Learned Sparse Retrieval - covering models, indexing structures, and evaluation resources.
A curated list of papers, code, models, and datasets for late-interaction multivector retrieval - covering ColBERT-style models, indexing techniques, software libraries, and benchmarks.
An extensive and commented list of resources on agentic search and agentic deep research - covering retrieval interfaces, the agentic search loop, RL-based training, systems, and benchmarks.
Official tusKANNy organization on Hugging Face, hosting models and datasets released alongside our research.
Used by the tools you already rely on
Come Find Us
We are based in Pisa, one of Italy's most storied university cities. Our team works across the University of Pisa and the National Research Council of Italy.
Whether you are a researcher exploring ANN, a practitioner seeking collaboration, or a company looking for tailored solutions, we would love to hear from you. Our doors in Pisa are always open.
info [AT] tuskanny [DOT] com
