# edgenext_small.usi_in1k vs mobilenetv3_small_100.lamb_in1k
Source: https://savrn.com/models/compare/edgenext-small-usi-in1k-vs-mobilenetv3-small-100-lamb-in1k
Markdown alternate of the page above; the site index is https://savrn.com/llms.txt

---

| Field | [edgenext_small.usi_in1k](https://savrn.com/models/edgenext-small-usi-in1k) timm/edgenext_small.usi_in1k | [mobilenetv3_small_100.lamb_in1k](https://savrn.com/models/mobilenetv3-small-100-lamb-in1k) timm/mobilenetv3_small_100.lamb_in1k |
| --- | --- | --- |
| Publisher | PyTorch Image Models | PyTorch Image Models |
| Task | Image classification | Image classification |
| Modality | Image | Image |
| Parameters, as reported | 6M parameters | 3M parameters |
| Architecture | Not stated | Not stated |
| Library | timm | timm |
| Context length | Not stated | Not stated |
| Repository size | 44.8 MB | 20.5 MB |
| Artifact formats | safetensors, pytorch | safetensors, pytorch |
| License | mit | apache-2.0 |
| Access | Open weights, no gate | Open weights, no gate |
| Memory at 16-bit (weights and margin) | 0 GB | 0 GB |
| Cheapest GPUs at 16-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Memory at 4-bit (weights and margin) | 0 GB | 0 GB |
| Cheapest GPUs at 4-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Revision viewed | 81bd8f2ee563 | 1824797e7887 |
| Downloads reported by the hub | 193.5k | 17.7M |
| Last observed | 2026-09-18 | 2026-10-23 |

An evaluation row appears only where at least two of these models report the same benchmark with the same stated configuration, metric, unit and setup. Different evaluators stay named in each cell. Values are shown as reported: no unit conversion, no ranking.

## SAVRN's Notes on edgenext_small.usi_in1k

EdgeNeXt is built for phones and edge devices, and this small version has about 6M parameters. It mixes convolutions with attention to get more accuracy per unit of compute, and the authors trained it on ImageNet-1K using distillation.

The MIT license allows commercial use. If you need an image classifier that runs on-device or on a CPU at high throughput, this is a sensible starting backbone to fine-tune. Measure latency on your actual target hardware, since edge performance varies more than parameter counts suggest.

SAVRN Research, 2026-09-18

## SAVRN's Notes on mobilenetv3_small_100.lamb_in1k

Every row of the memory table reads 0.0 GB, and that is the honest answer: 2,554,968 parameters and 20,541,585 bytes of weights for ImageNet-1k classification at 224×224, though the hub headline rounds the count to 3M. This is a CPU model; the one 192 GB MI300X at $1.85 an hour we list as its cheapest fit is the smallest unit a host rents, not a sizing answer. The 21.1M monthly downloads ride along with timm installs.

Apache 2.0 covers commercial use and redistribution of fine-tuned copies, notices kept. The 5 files include both bin and safetensors copies — take one. The card's training recipe matters more than any benchmark line: LAMB optimizer, EMA weight averaging, no CutMix, a step schedule run 50 percent longer than the ResNet Strikes Back A2 template. The page carries no reported evaluations, so pull accuracy from timm's published results and confirm the label set matches your images.

SAVRN Research

## Questions

### Which is larger, edgenext_small.usi_in1k or mobilenetv3_small_100.lamb_in1k?

edgenext_small.usi_in1k (6M parameters) is larger than mobilenetv3_small_100.lamb_in1k (3M parameters), by the parameter counts their publishers report.

### Which is cheaper to run, edgenext_small.usi_in1k or mobilenetv3_small_100.lamb_in1k?

At 4-bit, edgenext_small.usi_in1k fits on 1x MI300X from $1.85 an hour and mobilenetv3_small_100.lamb_in1k on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

### Can I use edgenext_small.usi_in1k commercially?

Yes. edgenext_small.usi_in1k is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.

### Can I use mobilenetv3_small_100.lamb_in1k commercially?

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.

## Related Comparisons

- [mobilenetv3_small_100.lamb_in1k vs mobilenetv4_conv_small.e2400_r224_in1k](https://savrn.com/models/compare/mobilenetv3-small-100-lamb-in1k-vs-mobilenetv4-conv-small-e2400-r224-in1k)
- [efficientnet_b0.ra_in1k vs mobilenetv3_small_100.lamb_in1k](https://savrn.com/models/compare/efficientnet-b0-ra-in1k-vs-mobilenetv3-small-100-lamb-in1k)