Output
: image embedding [1,768] from the attention pooler (unnormalized; L2-normalize before cosine)
Variants
All variants take and return fp32 tensors — swap the
.pte
file, keep your app code.
build
file
size (MB)
parity vs fp32 eager (worst corr)
Mac median (ms)*
fp32
siglip2_base_p16_image_xnnpack_fp32.pte
371.7
0.999994
195.6
fp16
siglip2_base_p16_image_xnnpack_fp16.pte
187.3
0.999986
227.8
Core ML (fp16, iOS)
siglip2_base_p16_image_coreml_all.pte
185.1
0.999806
4.4
The Core ML build is the same graph lowered to Apple's Neural Engine instead of
XNNPACK, which is CPU-only. Measured on an iPhone 17 Pro across seven models, it
runs
3.5x to 13.9x faster (median 12x)
at roughly half the file size — for
example Depth-Anything-V2-Small at 500.8 ms against 42.7 ms, and MODNet at 81.7 ms
against 5.9 ms. It computes in fp16 and is iOS-only; the XNNPACK files stay the
portable option and are what runs on Android.
*Mac arm64, single process, median of 10 — a reference point for relative cost
only, not a device number (torch eager fp32 on the same machine: 34.4 ms).
Verification (executorch 1.4.0, torch 2.13.0)
Parity is measured against the fp32 eager model on real image input;
corr
is
the correlation over all elements of each output tensor.
SigLIP2-base-patch16-224-ExecuTorch huggingface.co is an AI model on huggingface.co that provides SigLIP2-base-patch16-224-ExecuTorch's model effect (), which can be used instantly with this mlboydaisuke SigLIP2-base-patch16-224-ExecuTorch model. huggingface.co supports a free trial of the SigLIP2-base-patch16-224-ExecuTorch model, and also provides paid use of the SigLIP2-base-patch16-224-ExecuTorch. Support call SigLIP2-base-patch16-224-ExecuTorch model through api, including Node.js, Python, http.
SigLIP2-base-patch16-224-ExecuTorch huggingface.co is an online trial and call api platform, which integrates SigLIP2-base-patch16-224-ExecuTorch's modeling effects, including api services, and provides a free online trial of SigLIP2-base-patch16-224-ExecuTorch, you can try SigLIP2-base-patch16-224-ExecuTorch online for free by clicking the link below.
mlboydaisuke SigLIP2-base-patch16-224-ExecuTorch online free url in huggingface.co:
SigLIP2-base-patch16-224-ExecuTorch is an open source model from GitHub that offers a free installation service, and any user can find SigLIP2-base-patch16-224-ExecuTorch on GitHub to install. At the same time, huggingface.co provides the effect of SigLIP2-base-patch16-224-ExecuTorch install, users can directly use SigLIP2-base-patch16-224-ExecuTorch installed effect in huggingface.co for debugging and trial. It also supports api for free installation.
SigLIP2-base-patch16-224-ExecuTorch install url in huggingface.co: