Voxide segmentation models (ONNX)

ONNX exports of SAM 2.1 and SAM-Med3D (turbo) used by Voxide, a GPU volume viewer for microscopy, for interactive segmentation. Each model is an encoder/decoder pair run on the CPU with ONNX Runtime.

Voxide ships the SAM 2.1 Tiny pair. It downloads the others on request from this repository at a pinned revision and checks every file against the SHA-256 in SHA256SUMS, so a file that does not match is never loaded.

It also downloads five network-feature models for its pixel classifier (a DINOv2 ViT-S/14, a DINOv2 ViT-B/14, the first two VGG16 blocks, the Cellpose cyto3 backbone and micro-sam's ViT-T light-microscopy encoder), pinned and checked the same way: see Network-feature models for their files, provenance and licences.

And it downloads one spot-detection model for finding dots: Spotiflow's hybiss model, see Spot-detection model.

Files

Model Encoder Decoder Use in Voxide
SAM 2.1 Hiera Tiny sam2.1_hiera_tiny_encoder.onnx (110 MB) sam2.1_hiera_tiny_decoder.onnx (17 MB) 2D slices (default, fastest)
SAM 2.1 Hiera Small sam2.1_hiera_small_encoder.onnx (139 MB) sam2.1_hiera_small_decoder.onnx (17 MB) 2D slices
SAM 2.1 Hiera Base+ sam2.1_hiera_base_plus_encoder.onnx (278 MB) sam2.1_hiera_base_plus_decoder.onnx (17 MB) 2D slices
SAM 2.1 Hiera Large sam2.1_hiera_large_encoder.onnx (853 MB) sam2.1_hiera_large_decoder.onnx (17 MB) 2D slices (best masks, slowest)
SAM-Med3D turbo sammed3d_turbo_encoder.onnx (373 MB) sammed3d_turbo_decoder.onnx (31 MB) 3D volumes

Using them without the in-app download

Download the files you need and either use Models → Import model… in Voxide or copy them into Voxide's model folder. The file names must stay as they are: Voxide finds each model by its <name>_encoder.onnx / <name>_decoder.onnx pair.

pip install -U huggingface_hub
hf download voxide/voxide-models --include "sam2.1_hiera_small_*" --local-dir voxide-models

How they were made

Unmodified inference exports of the upstream checkpoints, produced with PyTorch 2.9.0 (ONNX opset 18, IR version 8). No fine-tuning, quantization or other change to the weights.

License and attribution

Both upstream model families are released under the Apache License 2.0 (see LICENSE), and so are these exports.

The network-feature models and the spot-detection model carry their own licences, so this card's licence is per file (license: other): Apache-2.0, BSD-3-Clause or CC-BY-4.0 (micro-sam's ViT-T encoder), as listed for each file in their section and in the spot-detection section, with the attribution each licence requires under Licences and attribution and Spotiflow's licence.

If you use these models in published work, please cite the original papers:

@article{ravi2024sam2,
  title   = {SAM 2: Segment Anything in Images and Videos},
  author  = {Ravi, Nikhila and others},
  journal = {arXiv preprint arXiv:2408.00714},
  year    = {2024}
}

@article{wang2023sammed3d,
  title   = {SAM-Med3D: Towards General-purpose Segmentation Models for Volumetric Medical Images},
  author  = {Wang, Haoyu and others},
  journal = {arXiv preprint arXiv:2310.15161},
  year    = {2023}
}

Network-feature models (pixel classifier)

Feature extractors for Voxide's pixel classifier: each turns an image tile into a per-pixel feature map that the classifier trains on. Single files, run on the CPU with ONNX Runtime; Voxide finds each by its exact file name, so keep the names as they are when importing them by hand.

Model File Size SHA-256 Licence
DINOv2 ViT-S/14 with registers dinov2-with-registers-small.onnx 88 522 426 bytes (89 MB) 815e440d222e60294bb5b165491c4760e7f08dc7ecd2488a5450e93d50887efe Apache-2.0
VGG16 blocks 1–2 vgg16-blocks12.onnx 1 042 858 bytes (1 MB) 230fb1431be9f9b6088fc68c24d4ff8c4894e7ef8ae3b16e14e0116b351f1b45 BSD-3-Clause
Cellpose cyto3 backbone cellpose-cyto3-backbone.onnx 26 483 022 bytes (26 MB) e689706a9709ba7fcf70480019bdd1a6a4a94eff05277dfd64d024f2d015195d BSD-3-Clause
DINOv2 ViT-B/14 with registers dinov2-with-registers-base.onnx 346 622 747 bytes (347 MB) 419f8937888b0d4cd607dccaed37b790b48d79cd1aad65a66bf96c7769e3b48b Apache-2.0
micro-sam ViT-T (light microscopy) microsam-vit-t-lm-encoder.onnx 27 983 702 bytes (28 MB) f4ee03d6c4e5286ee2a4caf92212ce2cc90349f9f26b0d45c3490e989dc814b3 CC-BY-4.0
hf download voxide/voxide-models --include "dinov2-*" "vgg16-*" "cellpose-*" --local-dir voxide-models
hf download voxide/voxide-models --include "microsam-*" --local-dir voxide-models

Provenance

  • DINOv2 ViT-S/14 with registers — Meta AI's facebook/dinov2-with-registers-small, in the fp32 ONNX export by onnx-community (onnx-community/dinov2-with-registers-small, file onnx/model.onnx at revision ec7abea1a8757ec4f9f7b26399d31234ffaa6e6a), copied here unchanged under a new name. Input pixel_values [N, 3, H, W] (H and W multiples of 14), output last_hidden_state; opset 14.
  • DINOv2 ViT-B/14 with registers — Meta AI's facebook/dinov2-with-registers-base, in the fp32 ONNX export by onnx-community (onnx-community/dinov2-with-registers-base, file onnx/model.onnx at revision fcd050203baf48e192055356f379b5949a111255), copied here unchanged under a new name. Input pixel_values [N, 3, H, W] (H and W multiples of 14), output last_hidden_state; opset 14. Voxide reads 768 features per 14 × 14 patch (768 × 32 × 32 for a 448 × 448 tile).
  • VGG16 blocks 1–2 — torchvision's ImageNet VGG16 (VGG16_Weights.IMAGENET1K_V1, vgg16-397923af.pth), cut after the second convolution block: the ReLU outputs of conv1_2 (64 channels) and of conv2_2 (128 channels, upsampled ×2 inside the graph), concatenated to a 192-channel stride-1 map. Exported by scripts/export_vgg16_onnx.py in the Voxide repository (pinned torch 2.9, torchvision 0.24, onnx 1.22; opset 17). The weights are unmodified.
  • Cellpose cyto3 backbone — Cellpose 3's built-in cyto3 model (cellpose 3.1.1), the U-Net's 32-channel stride-1 decoder output before the final 1×1 output convolution (no flows or cell probability). Input pixel_values [1, 2, H, W] (H and W multiples of 16; channel 0 the 1st–99th-percentile-normalised plane, channel 1 the nuclei channel or zeros). Exported by scripts/export_cellpose_onnx.py in the Voxide repository (pinned cellpose 3.1, onnx 1.22; opset 17). The weights are unmodified.
  • micro-sam ViT-T (light microscopy) — the image encoder of micro-sam's SAM LM Generalist ViT-T (vit_t_lm), from the BioImage Model Zoo record faithful-chicken v1.1, file vit_t.pt (SHA-256 32b92ab64959088878a1cc3e52b9b64aaa0ba4b24c98e9959746d7390bec7b7a). Exported by scripts/export_microsam_vit_t_onnx.py in the Voxide repository (pinned torch 2.14, onnx 1.22; opset 17): the image encoder alone (no prompt encoder or mask decoder), with a bilinear resize of the 448 × 448 input to 1024 × 1024 baked in front of it and its attention biases set to zero, as micro-sam's runtime has them. Input pixel_values [1, 3, 448, 448] (fixed; ImageNet-normalised RGB), output features [1, 256, 64, 64]: one 256-vector per 7 × 7 input pixels.

Licences and attribution

If you use these models in published work, please cite the original papers:

@article{oquab2023dinov2,
  title   = {DINOv2: Learning Robust Visual Features without Supervision},
  author  = {Oquab, Maxime and others},
  journal = {arXiv preprint arXiv:2304.07193},
  year    = {2023}
}

@article{darcet2023registers,
  title   = {Vision Transformers Need Registers},
  author  = {Darcet, Timoth{\'e}e and Oquab, Maxime and Mairal, Julien and Bojanowski, Piotr},
  journal = {arXiv preprint arXiv:2309.16588},
  year    = {2023}
}

@article{simonyan2014vgg,
  title   = {Very Deep Convolutional Networks for Large-Scale Image Recognition},
  author  = {Simonyan, Karen and Zisserman, Andrew},
  journal = {arXiv preprint arXiv:1409.1556},
  year    = {2014}
}

@article{stringer2021cellpose,
  title   = {Cellpose: a generalist algorithm for cellular segmentation},
  author  = {Stringer, Carsen and Wang, Tim and Michaelos, Michalis and Pachitariu, Marius},
  journal = {Nature Methods},
  volume  = {18},
  pages   = {100--106},
  year    = {2021}
}

@article{stringer2025cellpose3,
  title   = {Cellpose3: one-click image restoration for improved cellular segmentation},
  author  = {Stringer, Carsen and Pachitariu, Marius},
  journal = {Nature Methods},
  year    = {2025}
}

@article{archit2025microsam,
  title   = {Segment Anything for Microscopy},
  author  = {Archit, Anwai and Freckmann, Luca and Nair, Sushmita and others},
  journal = {Nature Methods},
  volume  = {22},
  pages   = {579--591},
  year    = {2025},
  doi     = {10.1038/s41592-024-02580-4}
}

Spot-detection model (dots)

Voxide's Find dots can use Spotiflow instead of its built-in method, per channel, on single planes. One file, run with ONNX Runtime on the CPU or, in a build with CUDA, on the graphics card; Voxide finds it by its exact file name.

Model File Size SHA-256 Licence
Spotiflow hybiss (2D) spotiflow-hybiss.onnx 37 903 970 bytes (38 MB) 90d6eabdd7eb00c2564af54d83e1e237cb132f9f564a405d18dbe2bb54663689 BSD-3-Clause
hf download voxide/voxide-models --include "spotiflow-*" --local-dir voxide-models

Provenance

Spotiflow's pretrained hybiss model (trained on HybISS images), from the spotiflow-models release 0.6.0 (hybiss.zip, MD5 254afa97c137d0bd74fd9c1827f0e323, checkpoint best.pt, SHA-256 fa5d5cb313bcfd75527f3d0962ccc3654bc7aa33359097d16f3b0835b0103bd1). Exported by scripts/export_spotiflow_onnx.py in the Voxide repository (pinned spotiflow 0.6.5, torch 2.15.0 CPU, onnx 1.23.2; opset 17): the network alone, unmodified weights. Input image [1, 1, H, W] (the plane scaled by its 1st and 99.8th percentiles, as Spotiflow does; H and W multiples of 16), outputs prob [1, 1, H, W] (the spot probability) and flow [1, 3, H, W] (the stereographic flow Voxide refines positions with). The export matches PyTorch within 1e-4, and Voxide's peak finding and refinement reproduce Spotiflow.predict's points.

Spotiflow's licence and attribution

Redistribution of the weights in ONNX form keeps the BSD-3-Clause notice above; the licence text is in the Spotiflow repositories linked. Please cite:

@article{dominguezmantes2025spotiflow,
  title   = {Spotiflow: accurate and efficient spot detection for fluorescence microscopy with deep stereographic flow regression},
  author  = {Dominguez Mantes, Albert and Herrera, Antonio and Khven, Irina and Schlaeppi, Anjalie and Kyriacou, Eftychia and Tsissios, Georgios and Skoufa, Evangelia and Santangeli, Luca and Buglakova, Elena and Durmus, Emine Berna and Manley, Suliana and Kreshuk, Anna and Arendt, Detlev and Aztekin, Can and Lingner, Joachim and La Manno, Gioele and Weigert, Martin},
  journal = {Nature Methods},
  year    = {2025},
  doi     = {10.1038/s41592-025-02662-x}
}
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for voxide/voxide-models

Quantized
(2)
this model

Papers for voxide/voxide-models