An Apple Core AI conversion of robbyant/lingbot-vision-vit-base — a vision encoder (ViT backbone) that maps an image to normalized per-patch feature tokens, for dense downstream tasks (depth, segmentation, spatial perception). Produced by coreai-fabric and indexed by coreai-catalog.

Feature backbone, not an end task. This is a frozen encoder: it emits per-patch feature tokens, not depths / masks / labels. The host owns image preprocessing (resize to the static size, ImageNet mean/std) and any downstream head. Use the upstream repo for the preprocessing + task heads.

Model facts

Field	Value
Parameters	86M
Architecture	transformer
Capabilities	image-feature-extraction
Image size	512px (static)
Patch size	16
Embed dim	768
Patch tokens	1024
Quantization / precision	none / float32
On-disk size	327 MB
Asset kind	single-graph ViT encoder (image -> per-patch tokens)
assetVersion	2.0

Use it — this needs host code you supply

The bundle is a single static-size graph: image [1,3,S,S] in → normalized patch_tokens [1, (S/16)^2, embed_dim] out. You supply the image preprocessing (resize to S, ImageNet normalize) and any downstream head in your host code (Swift or Python).

pip install coreai-catalog && coreai-catalog install lingbot-vision-vit-base

Requirements

Deployment: macOS 27.0+ / iOS 27.0+, Xcode 27+. The asset serializes with minimum_os v27, so the on-device Swift runtime requires macOS/iOS 27+. A Mac on macOS 26 can convert and inspect it but not run it on-device.
Apple Silicon.

Verification (output parity)

Gate A (structure): passed — the bundle's layout + metadata were validated; the graph loads.
Gate B — graph_output_cosine: 1.000000 min output cosine (median 1.000000) vs the fp32 torch backbone over 8 seeded images, measured on apple_silicon. Certifies the export computes the SAME per-patch tokens as the source backbone — a conversion-fidelity metric, not task accuracy.
This certifies the export is numerically faithful to the source backbone — it does NOT certify downstream task accuracy. Reproduce with coreai-fabric verify.

Provenance

Field	Value
Base model	robbyant/lingbot-vision-vit-base @ `f606f8c6c4002234ea68038f4d7c7cf57da96dfa`
Converted by	`models/lingbot/export.py` (version not reported)
Recipe	lingbot-vision-vit-base (recipe_source: fabric)
Precision / quantization	float32 / none
Conversion date	2026-07-07

Machine-readable, in this repo: parity-report.json · reproduce-manifest.json · LICENSE.

License and attribution

Weights licensed apache-2.0 — see the bundled LICENSE. This artifact is a converted derivative of the base backbone: its weights were converted to Apple Core AI format. The conversion itself is community work.

The on-device Core AI ecosystem

coreai-fabric — the reproducible recipe → .aimodel pipeline that produced this asset.
coreai-catalog — the index of Core AI models with provenance and integration snippets.
apple/coreai-models — Apple's official exporters and runtimes.

Not affiliated with Apple

Community conversion. Not produced, hosted, or endorsed by Apple. Apple and Core AI are trademarks of Apple Inc., used here only to describe the target runtime/format.

kevinqz/LingBot-Vision-ViT-Base-CoreAI

LingBot-Vision ViT-B (fabric)