Highlights

🦆 Lance x DuckDB: SQL for Retrieval on the Multimodal Lakehouse Format

The Lance extension for DuckDB turns DuckDB into a SQL compute engine over Lance datasets, exposing vector, full-text, and hybrid retrieval as SQL table functions. This enables fully composable retrieval workflows — joins with eval data, reproducible top-k slicing, SQL-based debugging, and materialization back into Lance.

🚗 Rethinking Table File Paths with Uber: Lance’s Multi-Base Layout

Working with Uber’s AI Infrastructure team, Lance introduced a multi-base layout to support product systems that need a single dataset to span multiple S3 buckets for parallel reads and writes.

📍 The Quest for One Million IOPS: Benchmarking Storage at Lance

Recent storage benchmarks in Lance reached up to 1.5 million IOPS by combining a scheduler rework with io_uring, showing that high random-access throughput depends more on reducing CPU overhead and context switching than on single-read latency.

Upcoming Events

February Open Data + AI Meetup - Peninsula, Bay Area Edition — Thursday, February 12

Hear from speakers from LanceDB, Fivetran, Dremio, and typedef about what they’re building and how they’re defining the future of open data and AI.

Register →

NYC Lakehouse Meetup — Tuesday, February 17

We’re bringing together Apache Iceberg, Lance, and Apache DataFusion communities in NYC to chat about all things open lakehouse and data infrastructure at Cloudflare’s NYC office.

Register →

Product Updates

LanceDB Enterprise Features

🔥 Add Page Cache Prewarm API

Users can prewarm LanceDB tables using a LanceDB administrative API. (It is also possible to prewarm some columns, but not others.)

🚦 Admission Control for Feature Engineering Jobs

Avoid deadlocks by rejecting jobs if the cluster does not have enough resources to execute the job.

⚙️ Adaptive Batch Sizing for Feature Engineering Job Checkpoints

Backfill jobs now change checkpoint size depending on udf execution time. Internal benchmarks show up to 2x performance improvements.

Open Source Updates

Lance and LanceDB Releases

Lance v1.0.4 (release notes)

🗂️ Multi-base storage layouts to span multiple buckets or regions with a single dataset
⚡ Faster query execution via tighter WAND bounds and reduced per-query overhead

LanceDB v0.28 (release notes)

🦆 DuckDB-native SQL retrieval for vector, FTS, and hybrid search
🧩 Expanded embedding support (VoyageAI v4, multimodal) and faster ingestion via parallel embedding computation

Lance-Graph v0.5.0 (release notes)

🕸️ Richer Cypher queries with WITH, COLLECT, & COUNT(DISTINCT …)
🔍 Vector search and similarity UDFs integrated directly into graph queries

Lance-Context v0.2.1 (release notes)

🧠 Versioned context store APIs for append, search, and checkout across Python and Rust
🗜️ Background compaction and reduced Python blocking for long-running systems

A huge thank you to contributors from Uber, Netflix, Hugging Face, Bytedance, Huawei, Tencent, Alibaba, and more for their contributions! 

Read the full newsletter for more updates around lance-namespace, lance-duckdb, lance-ray, and lance-spark.

🤝 Lance Community Sync Recap

In January, we held two Lance Community Syncs focused on the upcoming Lance v2.0.0 release, growing ecosystem integrations with DuckDB, Polaris, and Hugging Face, and the formalization of lance-context and lance-graph as official sub-projects.

The next Lance Community Sync will take place on February 12, 2026.

• Subscribe to the Lance mailing list to get the meeting invite →

• Add discussion topics to the meeting notes →

• Watch previous recordings: Jan 15 | Jan 29

ChanChan Mao

DevRel @ LanceDB

GitHub | LinkedIn