slide2vec

slide2vec encodes whole-slide images with publicly available pathology foundation models. It uses hs2p for tissue detection and tiling, and handles batching, multi-GPU execution, and embedding storage.

Install

Python 3.10 or newer is required:

pip install slide2vec

Many models need additional dependencies available through pip install "slide2vec[fm]". See the model installation guide for model-specific extras, separate environments, and upstream packages.

Reproducible model environments

The repository ships a uv.lock file that pins every package for each extra. From a checkout, uv sync --extra cu128 --extra fm installs the exact locked stack (use --extra cpu for CPU-only PyTorch). Add --extra flash-attn for PRISM2 and gigapath-slide. The Docker image is built from the same lock; its SLIDE2VEC_EXTRAS build argument selects the extras, for example --build-arg SLIDE2VEC_EXTRAS="cu128" for the base install only.

For gated models such as Virchow2, request access on the model's Hugging Face page and authenticate with hf auth login or an HF_TOKEN environment variable.

Embed a slide

from hs2p import TilingConfig
from slide2vec import Model, PreprocessingConfig

model = Model.from_preset("virchow2")
preprocessing = PreprocessingConfig(
    tiling=TilingConfig(
        requested_spacing_um=0.5, requested_tile_size_px=224, tolerance=0.05, overlap=0.0
    ),
)
embedded = model.embed_slide("/path/to/slide.svs", preprocessing=preprocessing)

tile_embeddings = embedded.tile_embeddings  # (N, 2560)
x, y = embedded.x, embedded.y               # level-0 tile coordinates

tiling is an hs2p TilingConfig. Leave it unset to take spacing and tile size from the preset; Virchow2 supports several scales, so it needs an explicit one. The preset also supplies precision. Use list_models() to list presets, or filter with list_models("tile"), list_models("slide"), or list_models("patient").

See getting started for preprocessing and execution settings, and the API guide for patient embeddings, image inputs, and dense grids.

Save a batch

Create a CSV manifest:

sample_id,image_path
slide-1,/data/slide-1.svs
slide-2,/data/slide-2.svs

Optional mask_path and spacing_at_level_0 columns supply a mask or correct missing or incorrect level-0 spacing. Patient-level models also require patient_id; see the manifest schema.

from slide2vec import ExecutionOptions, Model, Pipeline

pipeline = Pipeline(
    model=Model.from_preset("virchow2"),
    preprocessing=preprocessing,  # from the example above
    execution=ExecutionOptions(output_dir="outputs/run"),
)
result = pipeline.run(manifest_path="/path/to/slides.csv")

Runs use all available GPUs by default; set ExecutionOptions(num_gpus=2) to limit them. Embeddings are saved as .pt tensors with metadata sidecars. Use ExecutionOptions(output_format="npz") for NumPy archives. The output guide describes directories, shapes, coordinates, and progress records.

Add region_tile_multiple=6 to the preprocessing config to group tiles into 6×6 regions. These produce (num_regions, 36, feature_dim) tensors in hierarchical_embeddings/; see hierarchical features.

The same batch workflow is available from the terminal:

slide2vec /path/to/config.yaml

The CLI guide provides a complete config example, overrides, and resume instructions.

Docker

docker pull waticlems/slide2vec:latest
docker run --rm -it \
    -v /path/to/your/data:/data \
    -e HF_TOKEN \
    waticlems/slide2vec:latest

Set HF_TOKEN in your shell before starting the container.

More documentation

• Model zoo
• Performance benchmarks and QA
• API walkthrough notebook