Search public pages, research tools, and SAVRN solutions.
ResNet model pre-trained on ImageNet-1k at resolution 224x224. It was introduced in the paper Deep Residual Learning for Image Recognition by He et al.
What it takes to serve resnet-50 (26M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.0 GB | 0.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.0 GB | 0.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.
Sorting images into the ImageNet-1k classes is the job, and at 26M parameters it costs almost nothing in memory. Sixteen-bit weights take 0.1 GB, and the cheapest row in our runs_on table is one MI300X with 192 GB at $1.85 an hour. A card that size carries this classifier as a side load, so the question is how many images an hour justify holding any GPU for it.
Apache 2.0 permits commercial use, modification and redistribution, provided the license notices stay and significant changes are stated, with a patent grant from contributors. Access is open. Check that the ImageNet-1k labels match yours before committing, since that is the training set, and note the 410 MB on disk is stored in float32, heavier than what sits in memory. It ships as safetensors, pytorch, jax and tf; the residual design is in arXiv:1512.03385.
By Microsoft, published under apache-2.0, revision 34c2154c194f.
ResNet model pre-trained on ImageNet-1k at resolution 224x224. It was introduced in the paper Deep Residual Learning for Image Recognition by He et al.
Disclaimer: The team releasing ResNet did not write a model card for this model so this model card has been written by the Hugging Face team.
ResNet (Residual Network) is a convolutional neural network that democratized the concepts of residual learning and skip connections. This enables to train much deeper models.
This is ResNet v1.5, which differs from the original model: in the bottleneck blocks which require downsampling, v1 has stride = 2 in the first 1x1 convolution, whereas v1.5 has stride = 2 in the 3x3 convolution. This difference makes ResNet50 v1.5 slightly more accurate (\~0.5% top1) than v1, but comes with a small performance drawback (~5% imgs/sec) according to Nvidia.
8 files, 410.3 MB in total. The weights are 4 files totalling 410.3 MB in bin, h5, msgpack, safetensors.
| File | Type | Size | SHA-256 |
|---|---|---|---|
| flax_model.msgpack | Weights | 102.5 MB | 1dba0174a596 |
| model.safetensors | Weights | 102.5 MB | 9c6061af1f45 |
| pytorch_model.bin | Weights | 102.6 MB | ff8163a13233 |
| tf_model.h5 | Weights | 102.8 MB | 57afa67ca9f9 |
| config.json | Configuration | 69.6 KB | — |
| preprocessor_config.json | Configuration | 266 B | — |
| README.md | Documentation | 2.6 KB | — |
| .gitattributes | Repository | 1.2 KB | — |
Released by Microsoft through its official repository on Hugging Face. Read the license.
| Precision | Weights in memory |
|---|---|
| As published | 410.3 MB |
| 16-bit | 0.1 GB |
| 8-bit | 0.0 GB |
| 4-bit | 0.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
About 0.1 GB at 16-bit and 0 GB at 4-bit: the weights (26M parameters) plus a working margin. A long context needs more.
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Yes. resnet-50 is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
A ResNet-B image classification model. single layer 7x7 convolution with pooling 1x1 convolution shortcut downsample Trained on ImageNet-1k in timm using recipe template described below. ResNet Strikes Back A1 recipe - Deep Residual Learning for Image Recognition: https://arxiv.org/abs/1512.03385 Explore the dataset and runtime metrics of this model in timm model results.
A ResNet-B image classification model. single layer 7x7 convolution with pooling 1x1 convolution shortcut downsample Pretrained on Instagram-1B hashtags dataset using semi-weakly supervised learning and fine-tuned on ImageNet-1k by paper authors. - Billion-scale semi-supervised learning for image classification: https://arxiv.org/abs/1905.00546 - Deep Residual Learning for Image Recognition: https://arxiv.org/abs/1512.03385 Explore the dataset and runtime metrics of this model in timm model results.
A ResNet-D image classification model. 3-layer stem of 3x3 convolutions with pooling 2x2 average pool + 1x1 convolution shortcut downsample Trained on ImageNet-1k in timm using recipe template described below. RandAugment RA2 recipe. Inspired by and evolved from EfficientNet RandAugment recipes. Published as B recipe in ResNet Strikes Back. RMSProp (TF 1.0 behaviour) optimizer, EMA weight averaging Step (exponential decay w/ staircase) LR schedule with warmup - Deep Residual Learning for Image Recognition: https://arxiv.org/abs/1512.03385 - Bag of Tricks for Image Classification with Convolutional Neural Networks: https://arxiv.org/abs/1812.01187 Explore the dataset and runtime metrics of…
A ResNet-B image classification model. single layer 7x7 convolution with pooling 1x1 convolution shortcut downsample Trained on ImageNet-1k in timm using recipe template described below. Based on ResNet Strikes Back A1 recipe Stronger dropout, stochastic depth, and RandAugment than paper A1 recipe - Deep Residual Learning for Image Recognition: https://arxiv.org/abs/1512.03385 Explore the dataset and runtime metrics of this model in timm model results.
This model is a fine-tuned version of microsoft/resnet-50 on the imagefolder dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 0.0001 - trainbatchsize: 8 - evalbatchsize: 16 - gradientaccumulationsteps: 2 - totaltrainbatchsize: 16 - lrschedulertype: cosine - numepochs: 20 - labelsmoothingfactor: 0.1 - Transformers 5.16.1 - Pytorch 2.11.0+cu128 - Datasets 4.8.5 - Tokenizers 0.23.1
A Swin Transformer image classification model. Pretrained on ImageNet-1k by paper authors. Explore the dataset and runtime metrics of this model in timm model results.