pixano-inference-torch

PyTorch helpers shared by the torch-based Pixano Inference model packages (pixano-inference-sam, pixano-inference-clip, pixano-inference-grounding-dino, pixano-inference-transformers-vlm).

The Pixano Inference core is framework-free: it depends on no ML framework. A model package that runs on PyTorch depends on this package for device/dtype resolution (resolve_device, resolve_torch_dtype, should_compile), image-to-tensor conversion and NDArray/tensor bridging. torch is imported lazily, so importing these helpers during plugin discovery does not load torch.

Development

The package is self-contained, with its own lock file:

cd packages/pixano-inference-torch
uv sync
uv run pytest

Metadata

Release files for pixano-inference-torch 0.2.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for pixano-inference-torch 0.2.0
File	Size	Uploaded
pixano_inference_torch-0.2.0.tar.gz	53.9 kB	Oct 6, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for pixano-inference-torch 0.2.0
File	Interpreter	ABI	Platform	Reset
pixano_inference_torch-0.2.0-py3-none-any.whl 19.3 kB Oct 6, 2026	Python 3	none	any	Details

Total release size: 73.2 kB

Release files / pixano_inference_torch-0.2.0.tar.gz

Download URL	pixano_inference_torch-0.2.0.tar.gz
Size	53.9 kB
Tags	Source
SHA-256 checksum How to use checksums	`849f17aa79fffe335ac5328db4cb786d0b700ae3fdabad267ab5d6d5e251ec2e`
BLAKE2b-256 checksum How to use checksums	`14088d3feeba75c54dcfc99febe9c4e83fefb8189bd1329262f7b12a2ab87cd4`
Upload date	Oct 6, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/7.0.0 CPython/3.13.14`

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/pixano/pixano-inference
Publishing commit: github.com/pixano/pixano-inference/tree/6c637ff9d1231e605ed2256d64664c8bdccb6941

Workflow

Publishing configuration: github.com/pixano/pixano-inference/blob/6c637ff9d1231e605ed2256d64664c8bdccb6941/.github/workflows/publish.yml
Publishing logs: github.com/pixano/pixano-inference/actions/runs/37433720029/attempts/1

Artifact

Subject: pixano_inference_torch-0.2.0.tar.gz
SHA-256 checksum: 849f17aa79fffe335ac5328db4cb786d0b700ae3fdabad267ab5d6d5e251ec2e Verifying attestations

Release files / pixano_inference_torch-0.2.0-py3-none-any.whl

Download URL	pixano_inference_torch-0.2.0-py3-none-any.whl
Size	19.3 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`4f3166722446e4cb51e0258e103dfbdc538f103ef099286df4a5583faf2c42aa`
BLAKE2b-256 checksum How to use checksums	`a2a2dffd1b1234066945ea4905373fbbf5f18a5a3ffbf4145f882c7c30430cae`
Upload date	Oct 6, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/7.0.0 CPython/3.13.14`

Provenance

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/pixano/pixano-inference
Publishing commit: github.com/pixano/pixano-inference/tree/6c637ff9d1231e605ed2256d64664c8bdccb6941

Workflow

Publishing configuration: github.com/pixano/pixano-inference/blob/6c637ff9d1231e605ed2256d64664c8bdccb6941/.github/workflows/publish.yml
Publishing logs: github.com/pixano/pixano-inference/actions/runs/37433720029/attempts/1

Artifact

Subject: pixano_inference_torch-0.2.0-py3-none-any.whl
SHA-256 checksum: 4f3166722446e4cb51e0258e103dfbdc538f103ef099286df4a5583faf2c42aa Verifying attestations