File size: 1,491 Bytes
f12ad1d 0409e72 f12ad1d f66cc0f f12ad1d 0409e72 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 | ---
library_name: cellpose
pipeline_tag: image-segmentation
base_model: mouseland/cellpose-sam
base_model_relation: quantized
license: bsd-3-clause
arXiv: 2609.21038
tags:
- cellpose
- cellpose-sam
- microscopy
- quantization
- image-segmentation
---
<div align="center">
<img src="https://cdn-uploads.huggingface.co/production/uploads/68ed488e386d9944bc2ea3a5/z6lWF8anMTDnO-E3zrFTU.png" alt="Capicú Technologies" />
</div>
# Cellpose-SAM WQuant — W8A16
Ready-to-load `W8A16` quantized variant of
[Cellpose-SAM cpsam_v2](https://huggingface.co/mouseland/cellpose-sam).
## Load
```python
import runpy
from huggingface_hub import hf_hub_download
loader = runpy.run_path(
hf_hub_download("capicu-ai/cellpose-sam-wquant-w8a16", "runtime_loader.py")
)
model = loader["load_model"]()
```
Pass `device="cuda"` to `load_model` for CUDA.
## Result
- Weight size: 295.8 MiB
- Compression versus FP32: 3.93x
- Evaluated paired samples: 20
- Complete bounded-panel criterion met: Yes
- Reload verification normalized RMSE: 0.0
`results.json` and `model_registry.json` contain the compact machine-readable
record.
## Citation
If you use this model or code, please cite our paper:
```bibtex
@article{cruzromero2026retention,
title={Retention-Constrained Post-Training Quantization of Cellpose-SAM for Stem Cell Microscopy},
author={Cruz Romero, Sebasti{\'a}n A.},
journal={arXiv preprint arXiv:2609.21038},
note={Accepted at NeurIPS 2026 LXAI Workshop},
year={2026}
}
```
|