digit-cnn is an open-weight model from Abhi Agasanakoppa, released under MIT License. It has 431,530 parameters. At 16-bit it needs about 0 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 22 downloads a month.
Runs On
What it takes to serve digit-cnn (431,530 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 Oct 12, 2026.
digit-cnn on every accelerator the SAVRN Index prices, at every precision
Model Card
By Abhi Agasanakoppa, published under mit, revision c945f6dfe72d.
Read Abhi Agasanakoppa's full model card
license: mit datasets: - ylecun/mnist language: - en
Configuration
- Model type
- digit_cnn
Identity and Version
- Repository
- abhirath-a/digit-cnn
- Publisher
- Abhi Agasanakoppa
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 431,530 parameters
- Languages
- en
- Revision
- c945f6dfe72db3c8a2219fbfd9e8161c07350361
- First published
- 2026-08-24
- Last updated
- 2026-09-22
Files and Weights
6 files, 1.7 MB in total. The weights are 1 file totalling 1.7 MB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 1.7 MB | c01144b88c53 |
| config.json | Configuration | 272 B | — |
| inference.py | Configuration | 621 B | — |
| model.py | Configuration | 1.7 KB | — |
| README.md | Documentation | 60 B | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- mit
- Access
- Open weights, no gate
- Download size
- 1.7 MB
Released by Abhi Agasanakoppa through its official repository on Hugging Face. Read the license.
Built From
- Trained on (disclosed) ylecun/mnist
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 1.7 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.
Questions About digit-cnn
How much GPU memory does digit-cnn need?
About 0 GB at 16-bit and 0 GB at 4-bit: the weights (431,530 parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run digit-cnn 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 digit-cnn commercially?
Yes. digit-cnn 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.