Infrastructure / Storage & Retrieval

Quilt

Quilt is a scientific data management platform on AWS that packages data with metadata, documentation and version history for reuse. It ships a Python SDK and CLI for creating and installing versioned data packages on S3, and a self-deployed catalog stack (web UI, Lambda indexing services, and a search cluster) for browsing, searching and visualizing them.

quilt3 on PyPI is declared as the measurement identity, and the reasoning is recorded because the package is one of two components in this repository. Installing quilt3 gives you Quilt - the SDK and CLI in `api/python` create, version and install data packages against S3 on their own, with the catalog and Lambda stack in `catalog` and `lambdas` as an optional browsing layer over the same packages, all from the same Apache-2.0 source. Every programmatic use of Quilt goes through quilt3, so its downloads count Quilt users rather than clients of a separately shipped server, which is what separates it from weaviate-client or the openmldb SDK. The count is a lower bound (catalog-only users never install it) and the note says so. Verified 2026-09-02 via the GitHub API, the PyPI metadata (Repository points back at quiltdata/quilt) and the LICENSE body.

Openness

5 high confidence

5.0

license: Apache-2.0(OSI)
source: public
core-gated: ungated

Apache-2.0 license body confirmed. The single root LICENSE covers the whole repository - the Python SDK, the web catalog frontend and the Lambda backend services all in one tree - and the docs describe the "Enterprise Installation" as deploying this same repository's catalog and stack via CloudFormation into the customer's own AWS account, not a separately licensed or closed platform. The technical reference explicitly states the Lambda functions are open-source and independently security-audited. So source is public and the core ungated; the README's own "Enterprise platform" language describes a deployment and support model rather than functionality withheld from the published source.

https://github.com/quiltdata/quilt/blob/master/LICENSE recorded 2026-09-02
LICENSE file is the verbatim Apache-2.0 text, covering the entire repository with no per-directory carve-out.
https://api.github.com/repos/quiltdata/quilt recorded 2026-09-02
Repo metadata - license spdx_id Apache-2.0, private false, archived false, default branch master - for quiltdata/quilt.
https://github.com/quiltdata/quilt/blob/master/docs/Installation.md recorded 2026-09-02
Installation.md documents both the Python client (`pip install quilt3`) and the "Enterprise Installation" (the catalog and stack, deployed on AWS via CloudFormation from the Catalog Installation Guide) as installations of this same repository, with no license key or separate purchase gating the stack itself.

Adoption

2 medium confidence

2.0

32,910 PyPI downloads of `quilt3` in the trailing 30 days, which lands in the 10K-100K band of the software usage scale, level 2. quilt3 is the SDK and CLI built from this repository's `api/python`, and it creates, versions and installs Quilt packages against S3 without a catalog, so the count measures Quilt users rather than clients of a separate server. It is a lower bound - people who only browse a deployed catalog never install it - which is why confidence is medium. The stars scale would give the same level (1,370 stars, 1K-10K).

https://pypistats.org/api/packages/quilt3/recent recorded 2026-09-02
last_month downloads = 32,910 for quilt3.
https://api.github.com/repos/quiltdata/quilt recorded 2026-09-02
Repo metadata - stargazers_count = 1,370 - for quiltdata/quilt.

Capability

4 medium confidence

4.0

Banded on the category feature matrix as complete retrieval system, distributed or cluster-capable. Placed one band below the vespa anchor, alongside juicefs - Quilt provisions its own search cluster and indexing Lambdas as part of the deployed stack rather than versioning over storage it does not itself index, which is what keeps lakefs and dvc at the band below this one. It hosts no ranking or embedding model inside the query path, so it does not reach the top rung.

https://github.com/quiltdata/quilt/blob/master/docs/Installation.md recorded 2026-09-02
Enterprise Installation deploys the catalog and stack - including its own search cluster and indexing Lambdas - via CloudFormation from this repository's source.

Verified 2026-09-02