Downloads · 30 days
0
cmdr-clip-vit-b32-coreml
veszelovszki/cmdr-clip-vit-b32-coreml
cmdr-clip-vit-b32-coreml is a machine learning model from veszelovszki. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for coreml. The card lists the license as mit.
OpenAI's CLIP ViT-B/32 converted to Core ML (fp16) for on-device semantic image search in Cmdr, a fast, keyboard-first file manager for macOS.
Access
Public
Updated Jul 23, 2026
Repo size
475 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.zip475 MB · 100%
At a glance
- Library
- coreml
- License
- mit
- Access
- Public
- Created
- Jul 16, 2026
- Updated
- Jul 23, 2026
- SHA
- d75098e6
Base models
Model card
From the Hugging Face model README
CLIP ViT-B/32 for Core ML (Cmdr)
OpenAI's CLIP ViT-B/32 converted to Core ML (fp16) for on-device semantic image search in Cmdr, a fast, keyboard-first file manager for macOS.
Two towers, packaged as zipped .mlpackage bundles:
clip-image.mlpackage.zip: the image encoder (224×224 RGB → 512-dim embedding)clip-text.mlpackage.zip: the text encoder (77-token BPE ids → 512-dim embedding)
Conversion fidelity against the PyTorch reference: cosine similarity 1.0000 on the pinned reference vectors. The
conversion script and reference checks live in the Cmdr repo (apps/desktop/scripts/convert-clip-model/).
The original CLIP model is MIT-licensed by OpenAI; this conversion keeps that license.
- Library
- coreml
- License
- mit
- Created
- Jul 16, 2026
- Updated
- Jul 23, 2026