Skip to content

@lightonai LightOn

Change the repository type filter

All

    Repositories list

    50 repositories

    • next-plaid

      Public
      NextPlaid, ColGREP: Multi-vector search, from database to coding agents.
      cligrepvector-database
      cligrepvector-databaseagentic-ragmulti-vector
      Rust
      •
      Apache License 2.0
      •46•394•15•6•Updated Apr 24, 2026Apr 24, 2026

    • pylate

      Public
      Late Interaction Models Training & Retrieval
      information-retrievallanguage-modelrag
      information-retrievallanguage-modelragcolbert
      Python
      •
      MIT License
      •79•796•17•9•Updated Apr 23, 2026Apr 23, 2026

    • fast-plaid

      Public
      High-Performance Engine for Multi-Vector Search
      rustinformation-retrievalvector-database
      rustinformation-retrievalvector-databasecolbertcolpali
      Python
      •
      MIT License
      •21•248•6•2•Updated Apr 22, 2026Apr 22, 2026

    • demo-procurement-document-verification-DC4

      Public
      Demo LightOn API use case of a procurement document verification (DC4)
      Apache License 2.0
      •0•0•0•0•Updated Apr 21, 2026Apr 21, 2026

    • homebrew-tap

      Public
      Homebrew tap for LightOn tools
      Shell
      •0•0•0•0•Updated Apr 10, 2026Apr 10, 2026

    • BrowseComp-Plus

      Public
      BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent
      Python
      •42•2•0•0•Updated Mar 24, 2026Mar 24, 2026

    • fastkmeans-rs

      Public
      A Rust rewrite of FastKMeans for CPU-based clustering
      Rust
      •
      Apache License 2.0
      •1•14•0•0•Updated Mar 24, 2026Mar 24, 2026

    • bm25x

      Public
      A fast, streaming-friendly BM25 search engine in Rust with mmap support
      Rust
      •
      Apache License 2.0
      •3•48•1•0•Updated Mar 19, 2026Mar 19, 2026

    • homebrew-colgrep

      Public
      Homebrew tap for colgrep — semantic code search
      Ruby
      •0•1•0•0•Updated Feb 13, 2026Feb 13, 2026

    • distilabel

      Public
      Distilabel is a framework for synthetic data and AI feedback for engineers who need fast, reliable and scalable pipelines based on verified research papers.
      Python
      •
      Apache License 2.0
      •1•4•0•0•Updated Jan 30, 2026Jan 30, 2026

    • pylate-rs

      Public
      PyLate efficient inference engine
      pythonrustwasm
      pythonrustwasmcolbertlate-interaction
      Rust
      •
      MIT License
      •9•81•2•3•Updated Jan 7, 2026Jan 7, 2026

    • marenostrum-requirements

      Public
      Python
      •0•0•0•0•Updated Jan 6, 2026Jan 6, 2026

    • mt-rag-benchmark

      Public
      Multi-Turn RAG Benchmark
      Python
      •
      Apache License 2.0
      •29•0•0•0•Updated Sep 18, 2025Sep 18, 2025

    • ring-flash-attn

      Public
      Just here to get around some import issues with transformers. We need particular versions of transformers and it isn't compatible with the published package.
      Python
      •
      MIT License
      •0•1•0•0•Updated Aug 26, 2025Aug 26, 2025

    • sdk-python-paradigm

      Public
      Speakeasy generated python SDK for Paradigm
      0•0•0•0•Updated May 21, 2025May 21, 2025

    • trl

      Public
      TRL forked for RLVR
      Python
      •
      Apache License 2.0
      •2.7k•0•0•0•Updated Mar 20, 2025Mar 20, 2025

    • vllm-archive

      Public
      A high-throughput and memory-efficient inference and serving engine for LLMs
      Python
      •
      Apache License 2.0
      •16k•0•0•0•Updated Jan 24, 2025Jan 24, 2025

    • ducksearch

      Public
      Efficient BM25 with DuckDB 🦆
      information-retrievalbm25duckdb
      information-retrievalbm25duckdb
      Python
      •
      MIT License
      •2•66•0•0•Updated Dec 20, 2024Dec 20, 2024

    • .github

      Public
      0•0•0•0•Updated Sep 12, 2024Sep 12, 2024

    • torchtune

      Public
      Python
      •
      BSD 3-Clause "New" or "Revised" License
      •715•0•0•0•Updated Jul 5, 2024Jul 5, 2024

    • datatrove

      Public
      Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.
      Python
      •
      Apache License 2.0
      •254•3•0•0•Updated Jul 3, 2024Jul 3, 2024

    • composer

      Public
      Supercharge Your Model Training
      Python
      •
      Apache License 2.0
      •465•4•0•0•Updated Jun 20, 2024Jun 20, 2024

    • mamba-amd

      Public
      Port of Mamba to run and run efficiently on AMD.
      Python
      •
      Apache License 2.0
      •1.7k•5•0•0•Updated May 28, 2024May 28, 2024

    • causal-conv1d-amd

      Public
      Port of causal-conv1d to run and run efficiently on AMD.
      Cuda
      •
      BSD 3-Clause "New" or "Revised" License
      •178•3•0•0•Updated May 28, 2024May 28, 2024

    • text-embeddings-inference

      Public
      A blazing fast inference solution for text embeddings models
      Rust
      •
      Other
      •384•0•0•0•Updated Mar 25, 2024Mar 25, 2024

    • text-generation-inference

      Public
      Large Language Model Text Generation Inference
      Python
      •
      Other
      •1.3k•0•0•0•Updated Mar 18, 2024Mar 18, 2024

    • chroma

      Public
      the AI-native open-source embedding database
      Python
      •
      Apache License 2.0
      •2.2k•0•0•0•Updated Mar 4, 2024Mar 4, 2024

    • outlines

      Public archive
      Structured Text Generation
      Python
      •
      Apache License 2.0
      •690•0•0•0•Updated Mar 1, 2024Mar 1, 2024

    • opu-benchmarks

      Public archive
      ML benchmarks performance featuring LightOn's Optical Processing Unit (OPU) vs CPU and GPU.
      machine-learninghardware-accelerationphotonic-computing
      machine-learninghardware-accelerationphotonic-computing
      Python
      •0•23•0•4•Updated Jul 23, 2023Jul 23, 2023

    • transfer-learning-opu

      Public archive
      Optical Transfer Learning
      machine-learningdeep-learningtransfer-learning
      machine-learningdeep-learningtransfer-learningtensorrtphotonic-computing
      Jupyter Notebook
      •
      MIT License
      •3•27•0•4•Updated Jul 23, 2023Jul 23, 2023
    ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.