Models · Classification

CoAtNet

CoAtNet (Dai et al. 2021; timm's `rw` networks nano to 2, ImageNet-1k, some pretrained on ImageNet-12k): convolution stages followed by attention stages. Apache-2.0.

experimental Fork of huggingface/pytorch-image-models@0df212b · backends: portable, kernels

Checkpoint

Paper 2021 Added in 0.2.0 Evaluated with 0.2.0

Portable, RTX 5060 Ti 16 GB 7.23 ms 224x224, fp16
vs the faster original 1.1× original fp32: 7.88 ms
Peak memory 74 MiB original: 105 MiB
Accuracy vs original fp32 −0.02 portable top-1 %, ImageNet-1k val (82.58 original)

Use it

pip install fastforks

Lean install with only what this model needs: pip install "fastforks-core[timm]"

from fastforks import classification

model = classification.load("coatnet_nano_rw_224.sw_in1k")  # kernels on a supported GPU, else portable PyTorch
result = model(image)
result.logits
result.backend  # what ran: mode, device, dtype, kernels
fastforks bench coatnet_nano_rw_224.sw_in1k

Times the model on this machine. load(…, mode="portable") or mode="kernels" picks a backend; usage guide.

Speed and memory

Every backend on one GPU (RTX 5060 Ti 16 GB) and the same input, the authors’ code included. Kernels runs fp16. Portable runs fp16 on CUDA.

Input: synthetic, 224x224 (the task’s standard size)

BackenddtypeTime per call (ms)vs original fp32Peak memory
allocated / reserved
Load (s)First call (s)
Original fp32fp32 7.88–105 MiB/114 MiB1.60.36
Original fp16torch.autocast(float16) around the callfp16 10.60.75×124 MiB/146 MiB0.30.54
Portable fp32fp32not measured ––––
Portablefp16 7.231.1×74 MiB/86 MiB1.40.42
Kernelsnot measured ––––
Portable + torch.compile 4.711.7×96 MiB/126 MiB14 (build)–
TensorRT fp16fp16 0.70911×38 MiBenginenot comparable32 (build)–

Measured on RTX 5060 Ti 16 GB, torch 2.14.1+cu130, 2026-11-11. First call, 3 warm-up, peak reset, median of 3 rounds x median of 5. Ratios divide two times of this table (same GPU, same input). Original fp16 is the authors’ own precision option, else torch.autocast. torch.compile and TensorRT: build time instead of load time. TensorRT memory (engine, not comparable) is the GPU memory the engine took, read from the driver because TensorRT allocates outside PyTorch, so it can't be compared with the PyTorch peaks of the other rows.

Other devices

DeviceBackenddtypems / callPeak memoryOutput vs referenceInput
Apple M1 Pro 16 GB MPS, torch 2.14.1, 2026-11-14Originalfp3219.8 596 MiB process–224x224
Portablefp1619.6 572 MiB processmatches224x224
Portablefp3219.6 564 MiB processmatches224x224
Kernelsfp32not implemented–224x224

CUDA memory is torch’s peak allocated / reserved; Apple has no such counter, so the Mac rows show the process footprint (weights, Python and torch included). “Output vs reference” compares the device’s output with the stored golden output under the card’s tolerance; – = not checked on that device. Kernels “not implemented” = no kernels for this model on that backend yet: it runs portable there.

Licence and weights

Commercial use allowed

Code
Apache-2.0
Weights
Apache-2.0 source
Training data

Our reading of the licences, not legal advice.

Weights

coatnet_nano_rw_224.sw_in1k.safetensors 60.6 MB· hosted on Hugging Face

fastforks downloads and checks these files on first use (fastforks download coatnet_nano_rw_224.sw_in1k).

Accuracy

The test set through the authors’ code and through ours: same checkpoint, same protocol. The distance between the original’s fp32 and fp16 runs is what half precision alone costs; our runs should stay within it.

ImageNet-1k val

5000 random images, top-1 / top-5 %

Rundtypetop-1 % ↑top-5 % ↑Δ top-1 % vs fp32 originalimagesms / image
Original fp32fp3282.5895.82 reference 5000 1.59
Original fp16fp1682.5895.80 ±0.00 5000 0.970
Portable fp32fp3282.6095.78 +0.02 5000 2.05
Portablefp1682.5695.78 −0.02 5000 0.912
Kernelsnot measured

fastforks: the model's own resize / crop on the GPU; original: the library's pretrained loader and its PIL eval transform. Original = the authors’ code, same checkpoint and protocol. Measured on original fp32, original fp16: RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-11; portable fp32: RTX 5060 Ti, torch 2.14.1+cu130, fastforks 6f0a70e1, 2026-11-17; portable: RTX 5060 Ti, torch 2.14.1+cu130, fastforks 7ba7307f, 2026-11-11.

Closeness to the original

Beyond the test-set scores: the raw output of ours against the original’s fp32 output on the same input, pixel by pixel. CPU, float32: max |diff| <= 8.6e-06 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.0024 and p99 <= 0.011 (absolute error of the output logits).

CheckToleranceMeasured
CPU fp32, max |diff|≤ 0.0000086 not recorded
GPU default dtypes, mean |diff|≤ 0.0024 0.00121
GPU default dtypes, p99 |diff|≤ 0.011 0.0057

Recorded for coatnet_nano_rw_224.sw_in1k on RTX 5060 Ti, torch 2.10.0+cu130, 2026-11-17, with these kernels: cudnn_haccfvm_convfvm_extrafvm_mbfvm_r3

Details

CoAtNet (Dai et al. 2021) in timm's `rw` configurations: MBConv stages followed by transformer stages with relative position bias.

Input / output
Input: one RGB image. Output: ImageNet-1k class logits.
Input size
fixed 224x224: the input is resized (short side 248) and centre-cropped to it; tested up to 224x224 (74 MiB allocated)
Outputs
logits
Checkpoint
arch="coatnet_nano_rw_224.sw_in1k"
CUDA kernels
  • cudnn_haccconvolutions with fp16 accumulation (cuDNN)
  • fvm_convfused convolutions for image classifiers
  • fvm_extrafused transformer row operations (rotary split, SwiGLU with LayerNorm)
  • fvm_mbfused MBConv blocks (MobileNet and EfficientNet families)
  • fvm_r3further fused classifier blocks
Apple (MPS)
fp32, fp16
Full reference, generated from the repository

CoAtNet (Dai et al. 2021; timm's rw networks nano to 2, ImageNet-1k, some pretrained on ImageNet-12k): convolution stages followed by attention stages. Apache-2.0.

Commercial use: yes.

variantinputbackendtime (ms)peak memory (MiB, allocated / reserved)
coatnet_nano_rw_224.sw_in1ksynthetic (224x224)original (fp32)
original
portable
kernels
portable + torch.compile
TensorRT fp16
7.88
10.6
7.23
–
4.71
0.709
105 / 114
124 / 146
74 / 86
–
96 / 126
102 / 124
coatnet_rmlp_nano_rw_224.sw_in1ksynthetic (224x224)original (fp32)
original
portable
kernels
portable + torch.compile
TensorRT fp16
8.95
12.1
8.09
–
9.89
0.868
121 / 154
124 / 146
78 / 106
–
110 / 126
122 / 144
coatnet_0_rw_224.sw_in1ksynthetic (224x224)original (fp32)
original
portable
kernels
portable + torch.compile
TensorRT fp16
7.64
9.59
6.49
–
3.59
0.849
156 / 180
194 / 210
102 / 124
–
120 / 142
158 / 170
coatnet_1_rw_224.sw_in1ksynthetic (224x224)original (fp32)
original
portable
kernels
portable + torch.compile
TensorRT fp16
12.3
15.7
10.7
–
5.58
1.38
225 / 254
277 / 294
132 / 158
–
149 / 176
231 / 250
coatnet_rmlp_1_rw_224.sw_in1ksynthetic (224x224)original (fp32)
original
portable
kernels
portable + torch.compile
TensorRT fp16
14.0
18.4
12.4
–
6.54
1.47
213 / 234
278 / 296
131 / 156
–
155 / 176
243 / 262
coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1ksynthetic (224x224)original (fp32)
original
portable
kernels
portable + torch.compile
TensorRT fp16
13.9
18.3
12.3
–
6.63
1.55
226 / 254
278 / 296
132 / 158
–
162 / 176
257 / 276
coatnet_2_rw_224.sw_in12k_ft_in1ksynthetic (224x224)original (fp32)
original
portable
kernels
portable + torch.compile
TensorRT fp16
14.6
15.7
11.9
–
5.53
1.73
376 / 410
466 / 470
207 / 240
–
211 / 240
374 / 416
coatnet_rmlp_2_rw_224.sw_in1ksynthetic (224x224)original (fp32)
original
portable
kernels
portable + torch.compile
TensorRT fp16
15.3
18.6
15.1
–
6.34
1.92
376 / 410
466 / 472
207 / 240
–
238 / 260
404 / 446
coatnet_rmlp_2_rw_224.sw_in12k_ft_in1ksynthetic (224x224)original (fp32)
original
portable
kernels
portable + torch.compile
TensorRT fp16
15.3
18.6
12.8
–
6.52
1.87
376 / 410
466 / 470
207 / 240
–
238 / 260
402 / 444
coatnet_rmlp_2_rw_384.sw_in12k_ft_in1ksynthetic (384x384)original (fp32)
original
portable
kernels
portable + torch.compile
TensorRT fp16
27.9
17.9
13.9
–
6.92
3.69
520 / 566
516 / 528
296 / 330
–
326 / 370
464 / 500
Details
  • Original code: https://github.com/huggingface/pytorch-image-models at 0df212b369a5
  • Code licence: Apache-2.0 (permissive); files:
  • Weights licence: Apache-2.0
  • Licence evidence: https://huggingface.co/timm/coatnet_nano_rw_224.sw_in1k/blob/89fb9ccadc876a973c6ff954b089eb6723174c7a/README.md
  • Training data: ImageNet-1k (ILSVRC 2012 classification)
    • coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k: ImageNet-21k / ImageNet-22k (full ImageNet, fall 2011 release), ImageNet-1k (ILSVRC 2012 classification)
    • coatnet_2_rw_224.sw_in12k_ft_in1k: ImageNet-21k / ImageNet-22k (full ImageNet, fall 2011 release), ImageNet-1k (ILSVRC 2012 classification)
    • coatnet_rmlp_2_rw_224.sw_in12k_ft_in1k: ImageNet-21k / ImageNet-22k (full ImageNet, fall 2011 release), ImageNet-1k (ILSVRC 2012 classification)
    • coatnet_rmlp_2_rw_384.sw_in12k_ft_in1k: ImageNet-21k / ImageNet-22k (full ImageNet, fall 2011 release), ImageNet-1k (ILSVRC 2012 classification)
  • Kernel libraries: cudnn_hacc, fvm_conv, fvm_extra, fvm_mb, fvm_r3
  • Variants: coatnet_nano_rw_224.sw_in1k (default of coatnet_nano_rw_224), coatnet_rmlp_nano_rw_224.sw_in1k (default of coatnet_rmlp_nano_rw_224), coatnet_0_rw_224.sw_in1k (default of coatnet_0_rw_224), coatnet_1_rw_224.sw_in1k (default of coatnet_1_rw_224), coatnet_rmlp_1_rw_224.sw_in1k (default of coatnet_rmlp_1_rw_224), coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k (default of coatnet_rmlp_1_rw2_224), coatnet_2_rw_224.sw_in12k_ft_in1k (default of coatnet_2_rw_224), coatnet_rmlp_2_rw_224.sw_in1k, coatnet_rmlp_2_rw_224.sw_in12k_ft_in1k (default of coatnet_rmlp_2_rw_224), coatnet_rmlp_2_rw_384.sw_in12k_ft_in1k (default of coatnet_rmlp_2_rw_384)
  • Outputs: logits; label_set: imagenet-1k
  • Input size:
    • coatnet_nano_rw_224.sw_in1k, coatnet_rmlp_nano_rw_224.sw_in1k: fixed 224x224: the input is resized (short side 248) and centre-cropped to it
    • coatnet_0_rw_224.sw_in1k, coatnet_1_rw_224.sw_in1k, coatnet_rmlp_1_rw_224.sw_in1k, coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k, coatnet_2_rw_224.sw_in12k_ft_in1k, coatnet_rmlp_2_rw_224.sw_in1k, coatnet_rmlp_2_rw_224.sw_in12k_ft_in1k: fixed 224x224: the input is resized (short side 235) and centre-cropped to it
    • coatnet_rmlp_2_rw_384.sw_in12k_ft_in1k: fixed 384x384: the input is resized (short side 384) and centre-cropped to it
  • Tested up to (portable, fp16, RTX 5060 Ti 16 GB): coatnet_nano_rw_224.sw_in1k 224x224 (0.1 GiB peak allocated, 0.1 reserved); coatnet_rmlp_nano_rw_224.sw_in1k 224x224 (0.1 GiB peak allocated, 0.1 reserved); coatnet_0_rw_224.sw_in1k 224x224 (0.1 GiB peak allocated, 0.1 reserved); coatnet_1_rw_224.sw_in1k 224x224 (0.1 GiB peak allocated, 0.2 reserved); coatnet_rmlp_1_rw_224.sw_in1k 224x224 (0.1 GiB peak allocated, 0.2 reserved); coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k 224x224 (0.1 GiB peak allocated, 0.2 reserved); coatnet_2_rw_224.sw_in12k_ft_in1k 224x224 (0.2 GiB peak allocated, 0.2 reserved); coatnet_rmlp_2_rw_224.sw_in1k 224x224 (0.2 GiB peak allocated, 0.2 reserved); coatnet_rmlp_2_rw_224.sw_in12k_ft_in1k 224x224 (0.2 GiB peak allocated, 0.2 reserved); coatnet_rmlp_2_rw_384.sw_in12k_ft_in1k 384x384 (0.3 GiB peak allocated, 0.3 reserved)
  • coatnet_nano_rw_224.sw_in1k 224x224, first call: portable + torch.compile 14 s (cold Inductor cache), TensorRT fp16 32 s (ONNX + engine).
  • coatnet_rmlp_nano_rw_224.sw_in1k 224x224, first call: portable + torch.compile 14 s (cold Inductor cache), TensorRT fp16 37 s (ONNX + engine).
  • coatnet_0_rw_224.sw_in1k 224x224, first call: portable + torch.compile 12 s (cold Inductor cache), TensorRT fp16 38 s (ONNX + engine).
  • coatnet_1_rw_224.sw_in1k 224x224, first call: portable + torch.compile 16 s (cold Inductor cache), TensorRT fp16 42 s (ONNX + engine).
  • coatnet_rmlp_1_rw_224.sw_in1k 224x224, first call: portable + torch.compile 19 s (cold Inductor cache), TensorRT fp16 44 s (ONNX + engine).
  • coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k 224x224, first call: portable + torch.compile 18 s (cold Inductor cache), TensorRT fp16 43 s (ONNX + engine).
  • coatnet_2_rw_224.sw_in12k_ft_in1k 224x224, first call: portable + torch.compile 16 s (cold Inductor cache), TensorRT fp16 36 s (ONNX + engine).
  • coatnet_rmlp_2_rw_224.sw_in1k 224x224, first call: portable + torch.compile 18 s (cold Inductor cache), TensorRT fp16 46 s (ONNX + engine).
  • coatnet_rmlp_2_rw_224.sw_in12k_ft_in1k 224x224, first call: portable + torch.compile 18 s (cold Inductor cache), TensorRT fp16 41 s (ONNX + engine).
  • coatnet_rmlp_2_rw_384.sw_in12k_ft_in1k 384x384, first call: portable + torch.compile 19 s (cold Inductor cache), TensorRT fp16 42 s (ONNX + engine).
  • coatnet_nano_rw_224.sw_in1k original 224x224: torch.autocast(float16) around the call
  • coatnet_rmlp_nano_rw_224.sw_in1k original 224x224: torch.autocast(float16) around the call
  • coatnet_0_rw_224.sw_in1k original 224x224: torch.autocast(float16) around the call
  • coatnet_1_rw_224.sw_in1k original 224x224: torch.autocast(float16) around the call
  • coatnet_rmlp_1_rw_224.sw_in1k original 224x224: torch.autocast(float16) around the call
  • coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k original 224x224: torch.autocast(float16) around the call
  • coatnet_2_rw_224.sw_in12k_ft_in1k original 224x224: torch.autocast(float16) around the call
  • coatnet_rmlp_2_rw_224.sw_in1k original 224x224: torch.autocast(float16) around the call
  • coatnet_rmlp_2_rw_224.sw_in12k_ft_in1k original 224x224: torch.autocast(float16) around the call
  • coatnet_rmlp_2_rw_384.sw_in12k_ft_in1k original 384x384: torch.autocast(float16) around the call
  • coatnet_nano_rw_224.sw_in1k load / first call: original 0.28 s / 0.539 s, portable 1.37 s / 0.422 s
  • coatnet_rmlp_nano_rw_224.sw_in1k load / first call: original 0.26 s / 0.187 s, portable 1.36 s / 0.393 s
  • coatnet_0_rw_224.sw_in1k load / first call: original 0.38 s / 0.312 s, portable 1.37 s / 0.378 s
  • coatnet_1_rw_224.sw_in1k load / first call: original 0.47 s / 0.15 s, portable 1.44 s / 0.366 s
  • coatnet_rmlp_1_rw_224.sw_in1k load / first call: original 0.47 s / 0.022 s, portable 1.47 s / 0.373 s
  • coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k load / first call: original 0.47 s / 0.022 s, portable 1.49 s / 0.366 s
  • coatnet_2_rw_224.sw_in12k_ft_in1k load / first call: original 0.74 s / 0.314 s, portable 1.52 s / 0.505 s
  • coatnet_rmlp_2_rw_224.sw_in1k load / first call: original 0.71 s / 0.022 s, portable 1.54 s / 0.506 s
  • coatnet_rmlp_2_rw_224.sw_in12k_ft_in1k load / first call: original 0.71 s / 0.022 s, portable 1.8 s / 0.504 s
  • coatnet_rmlp_2_rw_384.sw_in12k_ft_in1k load / first call: original 0.74 s / 0.053 s (384x384), portable 1.52 s / 0.435 s (384x384)
  • Protocol, first call, 3 warm-up, peak reset, median of 3 rounds x median of 5

Measured

fastforks: the model's own resize / crop on the GPU; original: the library's pretrained loader and its PIL eval transform. Original = the original code named under Details, same checkpoint and protocol.

variantdatasetoriginal fp32original fp16portableportable fp32note
coatnet_nano_rw_224.sw_in1kImageNet-1k val, 5000 random images, top-1 / top-5 %82.58 / 95.8282.58 / 95.8082.56 / 95.7882.60 / 95.78
coatnet_rmlp_nano_rw_224.sw_in1kImageNet-1k val, 5000 random images, top-1 / top-5 %82.48 / 95.9882.54 / 95.9882.44 / 95.9682.40 / 95.96
coatnet_0_rw_224.sw_in1kImageNet-1k val, 5000 random images, top-1 / top-5 %83.34 / 96.3083.32 / 96.3083.38 / 96.3083.40 / 96.28
coatnet_1_rw_224.sw_in1kImageNet-1k val, 5000 random images, top-1 / top-5 %84.32 / 96.5484.38 / 96.5684.44 / 96.5684.46 / 96.56
coatnet_rmlp_1_rw_224.sw_in1kImageNet-1k val, 5000 random images, top-1 / top-5 %84.58 / 96.7484.58 / 96.7484.46 / 96.5684.50 / 96.56
coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1kImageNet-1k val, 5000 random images, top-1 / top-5 %85.34 / 97.0285.36 / 97.0285.30 / 97.0485.30 / 97.02
coatnet_2_rw_224.sw_in12k_ft_in1kImageNet-1k val, 5000 random images, top-1 / top-5 %87.04 / 98.1687.04 / 98.1487.18 / 98.1687.18 / 98.16
coatnet_rmlp_2_rw_224.sw_in1kImageNet-1k val, 5000 random images, top-1 / top-5 %85.12 / 96.8685.08 / 96.8685.16 / 96.8085.16 / 96.78
coatnet_rmlp_2_rw_224.sw_in12k_ft_in1kImageNet-1k val, 5000 random images, top-1 / top-5 %87.04 / 98.1887.02 / 98.1887.02 / 98.1287.02 / 98.12
coatnet_rmlp_2_rw_384.sw_in12k_ft_in1kImageNet-1k val, 5000 random images, top-1 / top-5 %87.72 / 98.6487.72 / 98.6887.84 / 98.5487.82 / 98.54

Measured on portable: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 7ba7307f, 2026-11-11; portable32: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 6f0a70e1, 2026-11-17.

Parity

CPU, float32: max |diff| <= 8.6e-06 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.0024 and p99 <= 0.011 (absolute error of the output logits).

variantkernels error vs original, mean / p99GPU, torchkernels used
coatnet_nano_rw_224.sw_in1k0.00121 / 0.0057RTX 5060 Ti, torch 2.10.0+cu130cudnn_hacc@cudnn91501, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120
coatnet_rmlp_nano_rw_224.sw_in1k0.00112 / 0.00488RTX 5060 Ti, torch 2.10.0+cu130cudnn_hacc@cudnn91501, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120
coatnet_0_rw_224.sw_in1k0.00169 / 0.00689RTX 5060 Ti, torch 2.10.0+cu130cudnn_hacc@cudnn91501, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120
coatnet_1_rw_224.sw_in1k0.00075 / 0.00309RTX 5060 Ti, torch 2.10.0+cu130cudnn_hacc@cudnn91501, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120
coatnet_rmlp_1_rw_224.sw_in1k0.00159 / 0.00642RTX 5060 Ti, torch 2.10.0+cu130cudnn_hacc@cudnn91501, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120
coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k0.00143 / 0.00522RTX 5060 Ti, torch 2.10.0+cu130cudnn_hacc@cudnn91501, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120
coatnet_2_rw_224.sw_in12k_ft_in1k0.00148 / 0.00543RTX 5060 Ti, torch 2.10.0+cu130cudnn_hacc@cudnn91501, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120
coatnet_rmlp_2_rw_224.sw_in1k0.00134 / 0.00632RTX 5060 Ti, torch 2.10.0+cu130cudnn_hacc@cudnn91501, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120
coatnet_rmlp_2_rw_224.sw_in12k_ft_in1k0.00144 / 0.0055RTX 5060 Ti, torch 2.10.0+cu130cudnn_hacc@cudnn91501, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120
coatnet_rmlp_2_rw_384.sw_in12k_ft_in1k0.00128 / 0.00505RTX 5060 Ti, torch 2.10.0+cu130cudnn_hacc@cudnn91501, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120

Example output

One fixed input through this checkpoint, the same for every model of the task (portable · float16 · NVIDIA GeForce RTX 5060 Ti).

Input

Top 5

  1. Egyptian cat 78%
  2. tiger cat 4.9%
  3. tabby 4.3%
  4. lynx 0.4%
  5. Siamese cat <0.1%
Chelsea the cat, input
Input · 100%

Input: Chelsea the cat, Stéfan van der Walt (CC0).