Skip to main content

xyberos-faiss

FAISS VectorStore plugin — RFC-0019, M4. A purely local, no-server vector store backed by FAISS. Passes the same parity scenarios as the stdlib SqliteVectorStore (the M4 Definition of Done).

Install

pip install -e ./faiss
pip install faiss-cpu             # or: pip install xyberos[vectors]

faiss-cpu wheels are available on PyPI for Linux/macOS; on Windows use a conda env or run the parity tests on a platform with wheels — the tests skip cleanly when the module is absent.

Usage

from xyberos import create_app
from xyberos_faiss import FaissPlugin

app = create_app()
app.load_plugin(FaissPlugin())

store = app.resolve("vector_store")
store.upsert("ns", "doc-1", [1.0, 0.0, 0.0, 0.0], {"text": "alpha"})
hits = store.query("ns", [1.0, 0.0, 0.0, 0.0], top_k=3)

Design

  • Each namespace is an in-memory faiss.IndexFlatIP over L2-normalized vectors, so inner-product scores equal cosine similarity (higher = more relevant, matching the VectorStore contract).
  • dim is optional and auto-detected from the first upsert (FAISS_DIM env honored).
  • faiss is imported lazily; a clear ProviderError is raised when it is missing.

Tests

pip install pytest
pytest tests/          # unit + parity + plugin tests

Parity tests run the exact scenarios the SqliteVectorStore reference passes; they skip cleanly when faiss-cpu is not importable on the current platform.

Ship location

[vectors] extra (faiss-cpu already added to pyproject.toml).

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

xyberos_faiss-0.1.0.tar.gz (6.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

xyberos_faiss-0.1.0-py3-none-any.whl (5.1 kB view details)

Uploaded Python 3

File details

Details for the file xyberos_faiss-0.1.0.tar.gz.

File metadata

  • Download URL: xyberos_faiss-0.1.0.tar.gz
  • Upload date:
  • Size: 6.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for xyberos_faiss-0.1.0.tar.gz
Algorithm Hash digest
SHA256 6b26c51d82163c1fde2e1c553464df2fe3fca1f4a698d4a1b5c964473b886ba5
MD5 841cf5d8121d1a5d2f57cbf476484aca
BLAKE2b-256 7993874afe6521086041ecc0c236492144408fb0d432822ce511c4ede3d01fc7

See more details on using hashes here.

Provenance

The following attestation bundles were made for xyberos_faiss-0.1.0.tar.gz:

Publisher: pypi-plugins.yml on xyberos/xyberos-plugins

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file xyberos_faiss-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: xyberos_faiss-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 5.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for xyberos_faiss-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 4a2aefcbe21fc20a93d69033499865280a66d25001e981586b36087e4646ad32
MD5 e2378080663d97d03ac926772af5dafc
BLAKE2b-256 36e592c17f4305033060111860b66fb8e5634f6f363a887cb4e54d3cafe46a90

See more details on using hashes here.

Provenance

The following attestation bundles were made for xyberos_faiss-0.1.0-py3-none-any.whl:

Publisher: pypi-plugins.yml on xyberos/xyberos-plugins

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

This release

0.1.0 This release

2 files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page