All variants take and return fp32 tensors — swap the
.pte
file, keep your app code.
precision
file
size (MB)
parity vs fp32 eager (worst corr)
Mac median (ms)*
fp32
real_esrgan_x4v3_xnnpack_fp32.pte
4.9
1.000000
33.9
*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: 42.3 ms).
Precisions that did not earn a slot
fp16 is not shipped
: it comes out at 100% of the fp32 file (4.9 MB vs 4.9 MB), so it buys nothing. XNNPACK serializes convolution weights as fp32 no matter what dtype the graph carries, so on a conv-heavy model fp16 saves no disk and only adds cast operations. Reach for int8 here, not fp16.
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.
output
shape
max_abs_diff
corr
0
[1, 3, 512, 512]
1.235e-05
1.000000
XNNPACK delegate coverage (fp32): 61.3% (106/173 ops); ops left on the portable kernels:
aten.gt.Scalar
x33,
aten.where.self
x33,
aten.upsample_nearest2d.vec
x1
Notes
: int8 does not run on executorch 1.4.0, and the reason is a collision between two workarounds rather than anything about this model. PReLU segfaults on the XNNPACK delegate (upstream #17559, fix #21480 not in the stable wheel), so it has to be excluded from partitioning — and a quantized graph with a portable PReLU between delegated convolutions fails shape propagation at execute. Three conv+PReLU layers reproduce it; the same graph in fp32 is corr 1.000000. Reported at
https://github.com/pytorch/executorch/pull/21480
Runs of mlboydaisuke Real-ESRGAN-x4v3-ExecuTorch on huggingface.co
47
Total runs
3
24-hour runs
11
3-day runs
21
7-day runs
21
30-day runs
More Information About Real-ESRGAN-x4v3-ExecuTorch huggingface.co Model
More Real-ESRGAN-x4v3-ExecuTorch license Visit here:
Real-ESRGAN-x4v3-ExecuTorch huggingface.co is an AI model on huggingface.co that provides Real-ESRGAN-x4v3-ExecuTorch's model effect (), which can be used instantly with this mlboydaisuke Real-ESRGAN-x4v3-ExecuTorch model. huggingface.co supports a free trial of the Real-ESRGAN-x4v3-ExecuTorch model, and also provides paid use of the Real-ESRGAN-x4v3-ExecuTorch. Support call Real-ESRGAN-x4v3-ExecuTorch model through api, including Node.js, Python, http.
Real-ESRGAN-x4v3-ExecuTorch huggingface.co is an online trial and call api platform, which integrates Real-ESRGAN-x4v3-ExecuTorch's modeling effects, including api services, and provides a free online trial of Real-ESRGAN-x4v3-ExecuTorch, you can try Real-ESRGAN-x4v3-ExecuTorch online for free by clicking the link below.
mlboydaisuke Real-ESRGAN-x4v3-ExecuTorch online free url in huggingface.co:
Real-ESRGAN-x4v3-ExecuTorch is an open source model from GitHub that offers a free installation service, and any user can find Real-ESRGAN-x4v3-ExecuTorch on GitHub to install. At the same time, huggingface.co provides the effect of Real-ESRGAN-x4v3-ExecuTorch install, users can directly use Real-ESRGAN-x4v3-ExecuTorch installed effect in huggingface.co for debugging and trial. It also supports api for free installation.
Real-ESRGAN-x4v3-ExecuTorch install url in huggingface.co: