Search public pages, research tools, and SAVRN solutions.
Open-weight model · Image classification
by PyTorch Image Models timm/mobilenetv3_small_100.lamb_in1k
A MobileNet-v3 image classification model. Trained on ImageNet-1k in timm using recipe template described below.
What it takes to serve mobilenetv3_small_100.lamb_in1k (3M 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 | 0.0 GB | 0.0 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.
Every row of the memory table for this model reads 0.0 GB. The weights total 20,541,585 bytes for 2,554,968 parameters, and the job is image classification over the ImageNet-1k label set. PyTorch Image Models trained it with a LAMB optimizer recipe similar to ResNet Strikes Back A2 but 50 percent longer, with EMA weight averaging and no CutMix. When the footprint rounds to zero at 16-bit, 8-bit and 4-bit, the cheapest configuration we list, a single 192 GB MI300X at $1.85 an hour, is not a sizing answer; the hardware question becomes images per hour and what else shares the card.
Apache 2.0 covers commercial use, modification and redistribution, so a fine-tuned version is yours to ship if you keep the notices and state what changed. Before committing, pull accuracy from timm's model results, since the file carries no reported evaluations, and confirm the ImageNet-1k classes match your images.
By PyTorch Image Models, published under apache-2.0, revision 1824797e7887.
A MobileNet-v3 image classification model. Trained on ImageNet-1k in timm using recipe template described below.
Recipe details:
* A LAMB optimizer based recipe that is similar to ResNet Strikes Back A2 but 50% longer with EMA weight averaging, no CutMix
* Step (exponential decay w/ staircase) LR schedule with warmup
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model('mobilenetv3_small_100.lamb_in1k', pretrained=True)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)
5 files, 20.5 MB in total. The weights are 2 files totalling 20.5 MB in bin, safetensors.
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 10.2 MB | 46d2c063b181 |
| pytorch_model.bin | Weights | 10.3 MB | c66054da8dbd |
| config.json | Configuration | 586 B | — |
| README.md | Documentation | 4.4 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
Released by PyTorch Image Models through its official repository on Hugging Face. Read the license.
| Precision | Weights in memory |
|---|---|
| As published | 20.5 MB |
| 16-bit | 0.0 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 GB at 16-bit and 0 GB at 4-bit: the weights (3M 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. mobilenetv3_small_100.lamb_in1k 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 MobileNet-V4 image classification model. Trained on ImageNet-1k by Ross Wightman. Trained with timm scripts using hyper-parameters inspired by the MobileNet-V4 paper with timm enhancements. NOTE: So far, these are the only known MNV4 weights. Official weights for Tensorflow models are unreleased. - MobileNetV4 -- Universal Models for the Mobile Ecosystem: https://arxiv.org/abs/2404.10518
A MobileNet-v3 image classification model. Trained on ImageNet-1k in Tensorflow by paper authors, ported to PyTorch by Ross Wightman. Explore the dataset and runtime metrics of this model in timm model results.
A very small test ResNet image classification model for testing and sanity checks. Trained on ImageNet-1k by Ross Wightman.
Swin Transformer v2 model pre-trained on ImageNet-1k at resolution 256x256. It was introduced in the paper Swin Transformer V2: Scaling Up Capacity and Resolution by Liu et al. and first released in this repository. Disclaimer: The team releasing Swin Transformer v2 did not write a model card for this model so this model card has been written by the Hugging Face team. The Swin Transformer is a type of Vision Transformer. It builds hierarchical feature maps by merging image patches (shown in gray) in deeper layers and has linear computation complexity to input image size due to computation of self-attention only within each local window (shown in red). It can thus serve as a general-purpose…
NOTE: Unless you are trying to detect imagery generated using older models such as VQGAN+CLIP, please use the updated version of this detector instead. This model is a proof-of-concept demonstration of using a ViT model to predict whether an artistic image was generated using AI. It was created in October 2022, and as such, the training data did not include any samples generated by Midjourney 5, SDXL, or DALLE-3. It still may be able to correctly identify samples from these more recent models due to being trained on outputs of their predecessors. Furthermore the intended scope of this tool is artistic images; that is to say, it is not a deepfake photo detector, and general computer imagery…
Trained on HAM10000. Educational project — not diagnostic-grade, not a medical device.