retrievalkit-embedding

Optional local FP32 text embeddings for RetrievalKit's Python API. This package is deliberately separate from the retrieval database distribution: constructing or prefetching an OnnxEmbedder may download verified model artifacts, while database initialization, indexing, search, and embedding inference never do.

from retrievalkit_embedding import OnnxEmbedder

OnnxEmbedder.prefetch()
embedder = OnnxEmbedder.load(local_only=True)
vector = embedder.embed("local retrieval")
assert len(vector) == 384

The default and only production-exposed model profile is the pinned FP32 all-MiniLM-L6-v2 export. Inputs use the qualified 256-token limit and every result contains exactly 384 finite, L2-normalized Python floats. This model precision is independent of RetrievalKit's default signed-I8 database storage.

Model artifacts are acquired only over verified HTTPS by the shared Rust ModelStore. Use cache_directory= to select an application cache and local_only=True to prohibit network access. Acquisition is visibly bounded to load(...) and prefetch(...); direct construction remains an equivalent convenience for compatibility.

The official ONNX Runtime 1.24.3 dynamic library must be supplied through runtime_library_path=, RETRIEVALKIT_ONNX_RUNTIME_LIBRARY, or the prepared package-local runtime directory. Package-local discovery accepts only the qualified macOS arm64 library with its pinned exact size and SHA-256. This repository does not contain the runtime binary.

This distribution is included in the v0.1.0 release inventory but is not available from PyPI until the protected release gates pass.

Qualification

The opt-in qualify.py utility is excluded from normal offline tests. It can emit model metadata and vectors for the shared conformance validator:

python qualify.py conformance --input texts.json --output python-fp32.json \
  --local-only

It also measures cached construction, the first public inference call, and 50 warm-ups followed by 750 measured 32-token single-text embeddings:

python qualify.py benchmark --local-only

Both commands require an installed extension, a qualified ONNX Runtime, and a downloaded cache when --local-only is selected. They never publish artifacts or packages.

Download files

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

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distributions

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

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

retrievalkit_embedding-0.1.0-cp314-cp314-macosx_11_0_arm64.whl (10.2 MB view details)

Uploaded Aug 2, 2026 CPython 3.14macOS 11.0+ ARM64

retrievalkit_embedding-0.1.0-cp313-cp313-macosx_11_0_arm64.whl (10.2 MB view details)

Uploaded Aug 2, 2026 CPython 3.13macOS 11.0+ ARM64

retrievalkit_embedding-0.1.0-cp312-cp312-macosx_11_0_arm64.whl (10.2 MB view details)

Uploaded Aug 2, 2026 CPython 3.12macOS 11.0+ ARM64

retrievalkit_embedding-0.1.0-cp311-cp311-macosx_11_0_arm64.whl (10.2 MB view details)

Uploaded Aug 2, 2026 CPython 3.11macOS 11.0+ ARM64

retrievalkit_embedding-0.1.0-cp310-cp310-macosx_11_0_arm64.whl (10.2 MB view details)

Uploaded Aug 2, 2026 CPython 3.10macOS 11.0+ ARM64

File details

Details for the file retrievalkit_embedding-0.1.0-cp314-cp314-macosx_11_0_arm64.whl.

File metadata

Download URL: retrievalkit_embedding-0.1.0-cp314-cp314-macosx_11_0_arm64.whl
Upload date: Aug 2, 2026
Size: 10.2 MB
Tags: CPython 3.14, macOS 11.0+ ARM64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for retrievalkit_embedding-0.1.0-cp314-cp314-macosx_11_0_arm64.whl
Algorithm	Hash digest
SHA256	`728ffd4e69feab1f4218e60221f39d686b4a9c123394a452925318d9e6a673e0`
MD5	`768c67c5978dc080fa8fdd8fe5486b0d`
BLAKE2b-256	`1ea87baa5360c7a7ec52e92215ced68c335732eda2a641fc7c8810be53c1b363`

See more details on using hashes here.

Provenance

The following attestation bundles were made for retrievalkit_embedding-0.1.0-cp314-cp314-macosx_11_0_arm64.whl:

Publisher: publish-release.yml on gungorbasa/RetrievalKit

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

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: retrievalkit_embedding-0.1.0-cp314-cp314-macosx_11_0_arm64.whl
- Subject digest: 728ffd4e69feab1f4218e60221f39d686b4a9c123394a452925318d9e6a673e0
- Sigstore transparency entry: 2318667175
- Sigstore integration time: Aug 2, 2026
Source repository:
- Permalink: gungorbasa/RetrievalKit@40fc6635deac814e6a20abc9aac8ac5ed0bd47bf
- Branch / Tag: refs/tags/v0.1.0-recovery.2
- Owner: https://github.com/gungorbasa
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish-release.yml@40fc6635deac814e6a20abc9aac8ac5ed0bd47bf
- Trigger Event: workflow_dispatch

File details

Details for the file retrievalkit_embedding-0.1.0-cp313-cp313-macosx_11_0_arm64.whl.

File metadata

Download URL: retrievalkit_embedding-0.1.0-cp313-cp313-macosx_11_0_arm64.whl
Upload date: Aug 2, 2026
Size: 10.2 MB
Tags: CPython 3.13, macOS 11.0+ ARM64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for retrievalkit_embedding-0.1.0-cp313-cp313-macosx_11_0_arm64.whl
Algorithm	Hash digest
SHA256	`7119e94e1092d2ac3bd69b267bee35105561face0ea70e5ebef2af648680031f`
MD5	`ecaba7cf7f1b31fa68e9a9197088b21f`
BLAKE2b-256	`f58218983fbb4124be946f4d6fbea3849d1037ef8f99bac5546c74eaf7168e24`

See more details on using hashes here.

Provenance

The following attestation bundles were made for retrievalkit_embedding-0.1.0-cp313-cp313-macosx_11_0_arm64.whl:

Publisher: publish-release.yml on gungorbasa/RetrievalKit

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

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: retrievalkit_embedding-0.1.0-cp313-cp313-macosx_11_0_arm64.whl
- Subject digest: 7119e94e1092d2ac3bd69b267bee35105561face0ea70e5ebef2af648680031f
- Sigstore transparency entry: 2318666939
- Sigstore integration time: Aug 2, 2026
Source repository:
- Permalink: gungorbasa/RetrievalKit@40fc6635deac814e6a20abc9aac8ac5ed0bd47bf
- Branch / Tag: refs/tags/v0.1.0-recovery.2
- Owner: https://github.com/gungorbasa
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish-release.yml@40fc6635deac814e6a20abc9aac8ac5ed0bd47bf
- Trigger Event: workflow_dispatch

File details

Details for the file retrievalkit_embedding-0.1.0-cp312-cp312-macosx_11_0_arm64.whl.

File metadata

Download URL: retrievalkit_embedding-0.1.0-cp312-cp312-macosx_11_0_arm64.whl
Upload date: Aug 2, 2026
Size: 10.2 MB
Tags: CPython 3.12, macOS 11.0+ ARM64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for retrievalkit_embedding-0.1.0-cp312-cp312-macosx_11_0_arm64.whl
Algorithm	Hash digest
SHA256	`5e9be3b4ef3f002ed7ea08fd9f3d3389019874960bac0b16e3525e1a1ba0c27a`
MD5	`dfd7f3aa3b9ecb51c8fea8e93b837564`
BLAKE2b-256	`fe476abfa79d6e988b2513fbd959675f78322c489666610aee63e7ee52075271`

See more details on using hashes here.

Provenance

The following attestation bundles were made for retrievalkit_embedding-0.1.0-cp312-cp312-macosx_11_0_arm64.whl:

Publisher: publish-release.yml on gungorbasa/RetrievalKit

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

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: retrievalkit_embedding-0.1.0-cp312-cp312-macosx_11_0_arm64.whl
- Subject digest: 5e9be3b4ef3f002ed7ea08fd9f3d3389019874960bac0b16e3525e1a1ba0c27a
- Sigstore transparency entry: 2318667244
- Sigstore integration time: Aug 2, 2026
Source repository:
- Permalink: gungorbasa/RetrievalKit@40fc6635deac814e6a20abc9aac8ac5ed0bd47bf
- Branch / Tag: refs/tags/v0.1.0-recovery.2
- Owner: https://github.com/gungorbasa
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish-release.yml@40fc6635deac814e6a20abc9aac8ac5ed0bd47bf
- Trigger Event: workflow_dispatch

File details

Details for the file retrievalkit_embedding-0.1.0-cp311-cp311-macosx_11_0_arm64.whl.

File metadata

Download URL: retrievalkit_embedding-0.1.0-cp311-cp311-macosx_11_0_arm64.whl
Upload date: Aug 2, 2026
Size: 10.2 MB
Tags: CPython 3.11, macOS 11.0+ ARM64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for retrievalkit_embedding-0.1.0-cp311-cp311-macosx_11_0_arm64.whl
Algorithm	Hash digest
SHA256	`b22810bd096eaaebc32a3463863310338658b44e8944e0011eff6a70e7a9499f`
MD5	`75d701171ead30abbba0c0ebb551026e`
BLAKE2b-256	`04c6844ac7577760457f595a1da26a7c8b367a1e0f58e137604eeec4024773b3`

See more details on using hashes here.

Provenance

The following attestation bundles were made for retrievalkit_embedding-0.1.0-cp311-cp311-macosx_11_0_arm64.whl:

Publisher: publish-release.yml on gungorbasa/RetrievalKit

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

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: retrievalkit_embedding-0.1.0-cp311-cp311-macosx_11_0_arm64.whl
- Subject digest: b22810bd096eaaebc32a3463863310338658b44e8944e0011eff6a70e7a9499f
- Sigstore transparency entry: 2318667016
- Sigstore integration time: Aug 2, 2026
Source repository:
- Permalink: gungorbasa/RetrievalKit@40fc6635deac814e6a20abc9aac8ac5ed0bd47bf
- Branch / Tag: refs/tags/v0.1.0-recovery.2
- Owner: https://github.com/gungorbasa
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish-release.yml@40fc6635deac814e6a20abc9aac8ac5ed0bd47bf
- Trigger Event: workflow_dispatch

File details

Details for the file retrievalkit_embedding-0.1.0-cp310-cp310-macosx_11_0_arm64.whl.

File metadata

Download URL: retrievalkit_embedding-0.1.0-cp310-cp310-macosx_11_0_arm64.whl
Upload date: Aug 2, 2026
Size: 10.2 MB
Tags: CPython 3.10, macOS 11.0+ ARM64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for retrievalkit_embedding-0.1.0-cp310-cp310-macosx_11_0_arm64.whl
Algorithm	Hash digest
SHA256	`d7228cd1dd54f191d250e73176cdc0c8b709ef53672515dcd5137c9bd18d7d9f`
MD5	`36c7ad7e651b0f035a1fdfaac2ad485e`
BLAKE2b-256	`c3440373f6b6809061abd20eb691b4a45b6c5349c9599a9a8a409e9366043bee`

See more details on using hashes here.

Provenance

The following attestation bundles were made for retrievalkit_embedding-0.1.0-cp310-cp310-macosx_11_0_arm64.whl:

Publisher: publish-release.yml on gungorbasa/RetrievalKit

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

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: retrievalkit_embedding-0.1.0-cp310-cp310-macosx_11_0_arm64.whl
- Subject digest: d7228cd1dd54f191d250e73176cdc0c8b709ef53672515dcd5137c9bd18d7d9f
- Sigstore transparency entry: 2318666872
- Sigstore integration time: Aug 2, 2026
Source repository:
- Permalink: gungorbasa/RetrievalKit@40fc6635deac814e6a20abc9aac8ac5ed0bd47bf
- Branch / Tag: refs/tags/v0.1.0-recovery.2
- Owner: https://github.com/gungorbasa
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish-release.yml@40fc6635deac814e6a20abc9aac8ac5ed0bd47bf
- Trigger Event: workflow_dispatch

Release history Release notifications | RSS feed

This release

0.1.0 This release

Aug 2, 2026 5 files

Pre-release

0.0.0a0

Aug 1, 2026 2 files

retrievalkit-embedding 0.1.0

retrievalkit-embedding

Qualification

Key dates

1 maintainer

Credits

License

Requires

Classifiers

Download files

Source Distributions

Built Distributions

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

Release history Release notifications | RSS feed