Skip to content
Search models or paste a model link. Press Enter to search.

Swin Tiny (224, s3)

timm/swin_s3_tiny_224.ms_in1k
Left and right arrow keys switch the vendor.
int8_embedl_tensorrt
timm swin_s3_tiny_224.ms_in1k architecture graph
timm swin_s3_tiny_224.ms_in1k architecture graph
  • timm/swin_s3_tiny_224.ms_in1k
  • INT8
  • Orin NX 16GB
Labels
  • Image
  • Classification
Build
Embedl-optimized
Precision
INT8
Runtime
TensorRT 10.16.2.10 with Embedl optimizations
Input size
224×224

swin_s3_tiny_224.ms_in1k is an image classifier from timm. It labels an image with one of the 1,000 ImageNet classes.

Files

Image Size

Model

docker.io/embedl/models:swin_s3_tiny_224.ms_in1k-timm_int8_embedl_tensorrt_orin-nx-16gb.2026102020125a90

—

Runtime

docker.io/embedl/models:runtimes.tensorrt.arm64.2026101815014ee0

—

Select hardware to start

Its measured speed and accuracy, its files and how to run it are per device.

Model variants (Pareto)

2 Embedl builds of Swin Tiny (224, s3) on 2 devices: latency against Top-1. Click a build to open it.

80.581.081.582.082.583.0Top-1 (%)123Latency (ms)↖ Better