# convnext_small.fb_in22k_ft_in1k_384 by PyTorch Image Models
Source: https://savrn.com/models/convnext-small-fb-in22k-ft-in1k-384
Markdown alternate of the page above; the site index is https://savrn.com/llms.txt

---

## Runs On

What it takes to serve convnext_small.fb_in22k_ft_in1k_384 (50M 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](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 |
| 8-bit | 0.1 GB | 0.1 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 |
| 4-bit | 0.0 GB | 0.0 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/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](https://savrn.com/ai-index/pricing/gpus), read Oct 22, 2026.

[convnext_small.fb_in22k_ft_in1k_384 on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/convnext-small-fb-in22k-ft-in1k-384/gpus)

## Model Card

By PyTorch Image Models, published under apache-2.0, revision 14a837353b6f.

### Model card for convnext_small.fb_in22k_ft_in1k_384

A ConvNeXt image classification model. Pretrained on ImageNet-22k and fine-tuned on ImageNet-1k by paper authors.

### Model Details

- Model Type: Image classification / feature backbone
- Model Stats:
- Params (M): 50.2
- GMACs: 25.6
- Activations (M): 63.4
- Image size: 384 x 384
- Papers:
- A ConvNet for the 2020s: https://arxiv.org/abs/2201.03545
- Original: https://github.com/facebookresearch/ConvNeXt
- Dataset: ImageNet-1k
- Pretrain Dataset: ImageNet-22k

### Model Usage

#### Image Classification

```
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('convnext_small.fb_in22k_ft_in1k_384', 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)
```

#### Feature Map Extraction

[Read the full model card (890 words)](https://savrn.com/models/convnext-small-fb-in22k-ft-in1k-384/card)

## Identity and Version

Repository

timm/convnext_small.fb_in22k_ft_in1k_384

Publisher

PyTorch Image Models

Task

Image classification

Modality

Image

Library

timm

Parameters

50M parameters

Languages

Not stated by the source

Revision

14a837353b6ffeced1c635dc5c8c799356e316f4

First published

2022-12-13

Last updated

2025-01-21

## Files and Weights

5 files, 402.0 MB in total. The weights are 2 files totalling 401.9 MB in bin, safetensors.

Weights2 files · 401.9 MB

Configuration1 file · 583 B

Documentation1 file · 15.7 KB

Repository1 file · 1.5 KB

Every file

| File | Type | Size | SHA-256 |
| --- | --- | --- | --- |
| model.safetensors | Weights | 200.9 MB | 17e4aa3c6ba4 |
| pytorch_model.bin | Weights | 201.0 MB | 24f8525364a6 |
| config.json | Configuration | 583 B | — |
| README.md | Documentation | 15.7 KB | — |
| .gitattributes | Repository | 1.5 KB | — |

## License and Download

License

apache-2.0

Access

Open weights, no gate

Download size

401.9 MB

[Download from PyTorch Image Models](https://huggingface.co/timm/convnext_small.fb_in22k_ft_in1k_384)

Released by PyTorch Image Models through its official repository on Hugging Face. [Read the license](https://www.apache.org/licenses/LICENSE-2.0).

## Built From

- Described by [arXiv:2201.03545](https://savrn.com/papers/a-convnet-for-the-2020s)
- Trained on (disclosed) imagenet-1k
- Trained on (disclosed) imagenet-22k

## Memory Requirements

| Precision | Weights in memory |
| --- | --- |
| As published | 401.9 MB |
| 16-bit | 0.1 GB |
| 8-bit | 0.1 GB |
| 4-bit | 0.0 GB |

Weights only, from the published parameter count; the key-value cache and runtime add to this.

## Questions About convnext_small.fb_in22k_ft_in1k_384

### How much GPU memory does convnext_small.fb_in22k_ft_in1k_384 need?

About 0.1 GB at 16-bit and 0 GB at 4-bit: the weights (50M parameters) plus a working margin. A long context needs more.

### What is the cheapest GPU to run convnext_small.fb_in22k_ft_in1k_384 on?

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.

### Can I use convnext_small.fb_in22k_ft_in1k_384 commercially?

Yes. convnext_small.fb_in22k_ft_in1k_384 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.

## Similar Models

Model · Image classification

### [convnext_small.in12k_ft_in1k](https://savrn.com/models/convnext-small-in12k-ft-in1k)

[PyTorch Image Models](https://savrn.com/model-publishers/timm)

A ConvNeXt image classification model. Pretrained in timm on ImageNet-12k (a 11821 class subset of full ImageNet-22k) and fine-tuned on ImageNet-1k by Ross Wightman. ImageNet-12k training done on TPUs thanks to support of the TRC program. Fine-tuning performed on 8x GPU Lambda Labs cloud instances. Explore the dataset and runtime metrics of this model in timm model results. All timing numbers from eager model PyTorch 1.13 on RTX 3090 w/ AMP.

Open weights apache-2.0 50M parameters timm

[View model](https://savrn.com/models/convnext-small-in12k-ft-in1k)

Model · Image classification

### [cspnext_x.rsb_a1_in1k](https://savrn.com/models/cspnext-x-rsb-a1-in1k)

[Munehiro Kobayashi](https://savrn.com/model-publishers/munehiro-k)

A CSPNeXt image classification model. Pretrained on ImageNet-1k by OpenMMLab (RTMDet) and converted to timm format. Name disambiguation: This is the CSPNeXt backbone of RTMDet (OpenMMLab), as implemented in MMDetection. A separate paper, CSPNeXt: A new efficient token hybrid backbone (Chen et al., EAAI 2024, doi:10.1016/j.engappai.2024.107886), uses the same name for a different architecture. These weights do not implement that paper; please cite RTMDet (below) for this model. - https://github.com/open-mmlab/mmdetection/tree/main/configs/rtmdet/classification - https://download.openmmlab.com/mmdetection/v3.0/rtmdet/cspnextrsbpretrain/cspnext-x8xb256-rsb-a1-600ein1k-b3f78edd.pth ImageNet-1k…

Open weights apache-2.0 49M parameters timm

[View model](https://savrn.com/models/cspnext-x-rsb-a1-in1k)

Model · Image classification

### [PlumaDINO-vits16plus-preview](https://savrn.com/models/plumadino-vits16plus-preview)

[YX](https://savrn.com/model-publishers/yxsi)

PlumaDINO-vits16plus-preview is an independently fine-tuned image classifier for living bird species, built on the distilled DINOv3 ViT-S+/16 backbone. This preview supports aspect-ratio-preserving inputs and loads directly with Transformers. Install the dependencies in requirements.txt in your own environment. Run this example from the parent of the downloaded model folder. It uses CPU; set device = "cuda" to use a GPU. After publication, a Hub repository ID may replace the local folder path. For an offline local copy, pass localfilesonly=True to both loaders. AutoModel.frompretrained(..., trustremotecode=True) also loads the classifier. The folder includes the custom inference code; a…

Open weights other 38M parameters transformers

[View model](https://savrn.com/models/plumadino-vits16plus-preview)

Model · Image classification

### [wide_resnet50_2.racm_in1k](https://savrn.com/models/wide-resnet50-2-racm-in1k)

[PyTorch Image Models](https://savrn.com/model-publishers/timm)

A Wide-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. RandAugment RACM 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 Explore the dataset and runtime metrics of this model in timm model results.

Open weights apache-2.0 69M parameters timm

[View model](https://savrn.com/models/wide-resnet50-2-racm-in1k)

Model · Image classification

### [convnextv2_tiny.fcmae_ft_in22k_in1k](https://savrn.com/models/convnextv2-tiny-fcmae-ft-in22k-in1k)

[PyTorch Image Models](https://savrn.com/model-publishers/timm)

A ConvNeXt-V2 image classification model. Pretrained with a fully convolutional masked autoencoder framework (FCMAE) and fine-tuned on ImageNet-22k and then ImageNet-1k. Explore the dataset and runtime metrics of this model in timm model results. All timing numbers from eager model PyTorch 1.13 on RTX 3090 w/ AMP.

Open weights cc-by-nc-4.0 29M parameters timm

[View model](https://savrn.com/models/convnextv2-tiny-fcmae-ft-in22k-in1k)

Model · Image classification

### [convnextv2-tiny-22k-224](https://savrn.com/models/convnextv2-tiny-22k-224)

[AI at Meta](https://savrn.com/model-publishers/facebook)

ConvNeXt V2 model pretrained using the FCMAE framework and fine-tuned on the ImageNet-22K dataset at resolution 224x224. It was introduced in the paper ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders by Woo et al. and first released in this repository. Disclaimer: The team releasing ConvNeXT V2 did not write a model card for this model so this model card has been written by the Hugging Face team. ConvNeXt V2 is a pure convolutional model (ConvNet) that introduces a fully convolutional masked autoencoder framework (FCMAE) and a new Global Response Normalization (GRN) layer to ConvNeXt. ConvNeXt V2 significantly improves the performance of pure ConvNets on various…

Open weights apache-2.0 29M parameters transformers

[View model](https://savrn.com/models/convnextv2-tiny-22k-224)

## PyTorch Image Models

[All models and datasets](https://savrn.com/model-publishers/timm)

## Versions

- [14a837353b6f](https://savrn.com/models/convnext-small-fb-in22k-ft-in1k-384/versions/14a837353b6f) · current 2026-09-18

## Explore More

- [All image classification models](https://savrn.com/models/tasks/image-classification)
- [All models under apache-2.0](https://savrn.com/models/licenses/apache-2-0)
- [Model comparisons](https://savrn.com/models/comparisons)
- [The model directory](https://savrn.com/models)
- [Open model prices by host](https://savrn.com/ai-index/pricing/open-models)

## Source

- Repository metadata, read 2026-10-20.
- [PyTorch Image Models website](https://github.com/huggingface/pytorch-image-models)
- [Hugging Face record](https://huggingface.co/timm/convnext_small.fb_in22k_ft_in1k_384)
- [How the hub is built](https://savrn.com/model-hub/methodology)