Help preserve open and free AI for humanity's future

← All models

CIDAS_clipseg-rd64-refined

CIDAS · View on Hugging Face ↗

Zero-shot image segmentation from a text prompt (CLIPSeg, refined weights) — segment anything you can name.

✓ verified · rehash-vs-hf-metadata at 2026-08-24T11:39:58Z

apache-2.01.12 GB (1,207,784,944 bytes)transformerspytorchsafetensorsclipsegvisionimage-segmentationpaper: 2112.10003

Get this model

Download CIDAS_clipseg-rd64-refined.torrent

Recommended — the .torrent carries the webseed url-list, so your client can fall back to plain HTTPS if the swarm is thin. See/verify for the full download + verification walkthrough.

Model card

The complete upstream card, rendered from this payload's README.md — the same hash-verified bytes the torrent distributes. Images and off-site links are removed; the original card on Hugging Face carries them.


license: apache-2.0 tags:

  • vision
  • image-segmentation inference: false

CLIPSeg model

CLIPSeg model with reduce dimension 64, refined (using a more complex convolution). It was introduced in the paper Image Segmentation Using Text and Image Prompts by Lüddecke et al. and first released in this repository.

Intended use cases

This model is intended for zero-shot and one-shot image segmentation.

Usage

Refer to the documentation.

Magnet link (secondary — no webseeds)

Opens the swarm directly, but carries no webseed url-list. Prefer the.torrent download above — HTTP fallback seeds ride inside it.

magnet:?xt=urn:btih:53dcd479fef2d7566eb016f4d3ae333de00e5243&dn=CIDAS_clipseg-rd64-refined

Open magnet in torrent client · infohash 53dcd479fef2d7566eb016f4d3ae333de00e5243

Files & hashes

PathSizeMethodHash
README.md596 B (596 B)sha1-git-blobbdb1f87160d83a7fc2aa6a4837f09a17546ea099
config.json4.6 KB (4,732 B)sha1-git-blob9c1a3a8693f29070857102aea691cadcb7e8a806
merges.txt512.3 KB (524,619 B)sha1-git-blob76e821f1b6f0a9709293c3b6b51ed90980b3166b
model.safetensors575.1 MB (603,049,496 B)sha256-lfsd00ca85d6b859f9d07b7cfb8ef26fe9771cb275b34c9368f2ecf603139307f55
preprocessor_config.json380 B (380 B)sha1-git-blob30fe86be4cc83725316cbc68ed8c78f9d1483642
pytorch_model.bin575.2 MB (603,143,713 B)sha256-lfsdd9308225b8314bb7236f207e6ea72b22db5d90dba03fe3dc7d654f54dcfd08a
special_tokens_map.json472 B (472 B)sha1-git-blob2c2130b544c0c5a72d5d00da071ba130a9800fb2
tokenizer_config.json974 B (974 B)sha1-git-blob446c39016ef7112b9ace812c98641502da0193d1
vocab.json1.0 MB (1,059,962 B)sha1-git-blob469be27c5c010538f845f518c4f5e8574c78f7c8

Provenance

Upstream repositoryCIDAS/clipseg-rd64-refined
Revision (pinned)999e0328d9e10b484360c477313983f9afdd7050
Fetched at2026-08-24T11:39:23Z
License at fetchapache-2.0
Snapshot toolhuggingface · seedbank 0.1.0

Trackers

Webseeds