<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
    <title>lance</title>
    <link rel="self" type="application/atom+xml" href="https://links.biapy.com/guest/tags/3296/feed"/>
    <updated>2026-08-07T21:20:00+00:00</updated>
    <id>https://links.biapy.com/guest/tags/3296/feed</id>
            <entry>
            <id>https://links.biapy.com/links/12769</id>
            <title type="text"><![CDATA[Firn]]></title>
            <link rel="alternate" href="https://firnflow.io/" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/12769"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[Multi-tenant vector search on object storage.
Vector search on object storage.

 The cost efficiency of S3 with the speed of local RAM. A multi-tenant vector and full-text search engine featuring a tiered RAM → NVMe → S3 architecture for microsecond latency on top of object storage 

A multi-tenant search engine that pairs LanceDB with a tiered RAM + NVMe cache. Every namespace lives on cheap object storage (AWS S3, MinIO, R2, Tigris, DigitalOcean Spaces, or native Google Cloud Storage) with near-zero idle cost. Every cache hit costs zero backend requests. 

- [Firn @ GitHub](https://github.com/gordonmurray/firnflow).

Related contents:

- [S3 is the perfect place to store data, until you try to search it @ Gordon Murray](https://gordonmurray.ie/data/2026/05/02/s3-is-the-perfect-place-to-store-data-until-you-try-to-search-it.html).]]>
            </summary>
            <updated>2026-05-15T14:07:08+00:00</updated>
        </entry>
            <entry>
            <id>https://links.biapy.com/links/10697</id>
            <title type="text"><![CDATA[Lance]]></title>
            <link rel="alternate" href="https://lancedb.github.io/lance/" />
            <link rel="via" type="application/atom+xml" href="https://links.biapy.com/links/10697"/>
            <author>
                <name><![CDATA[Biapy]]></name>
            </author>
            <summary type="text">
                <![CDATA[Modern columnar data format for ML and LLMs implemented in Rust. Convert from parquet in 2 lines of code for 100x faster random access, vector index, and data versioning. Compatible with Pandas, DuckDB, Polars, Pyarrow, and PyTorch with more integrations coming.. 

Lance is a modern columnar data format optimized for machine learning and AI applications. It efficiently handles diverse multimodal data types while providing high-performance querying and versioning capabilities.

- [Lance @ GitHub](https://github.com/lancedb/lance).

Related contents:

- [Lance takes aim at Parquet in file format joust @ The Register](https://www.theregister.com/2025/10/14/lance_parquet/).]]>
            </summary>
            <updated>2025-10-17T12:01:52+00:00</updated>
        </entry>
    </feed>
