classification-proto is an open-weight model from Leandro.costa, released under BSD 3-Clause License. It has 16,576 parameters and a 512-token context. 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.
This is an experimental Cnn Transformer codebase for Classification. It keeps the giant setup intentionally manageable so architecture changes can be inspected before a full training run.
Runs On
What it takes to serve classification-proto (16,576 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 18, 2026.
classification-proto on every accelerator the SAVRN Index prices, at every precision
Model Card
By Leandro.costa, published under bsd-3-clause, revision fa976936e5a4.
This is an experimental Cnn Transformer codebase for Classification. It keeps the giant setup intentionally manageable so architecture changes can be inspected before a full training run. - The Python file contains the model and runnable example or training entry point. - config.json records the generated architecture settings. - trainingargs.json records the default experiment recipe. - model.safetensors is a valid initialization checkpoint for smoke tests; it is not presented as a trained benchmark checkpoint. - No benchmark score is claimed in this repository. The included configuration uses novograd with a constant warmup schedule. These are starting values in the script, not evidence…
Read Leandro.costa's full model card
Cnn Transformer for Classification
Overview
This is an experimental Cnn Transformer codebase for Classification. It keeps the giant setup intentionally manageable so architecture changes can be inspected before a full training run.
Repository status
- The Python file contains the model and runnable example or training entry point.
config.jsonrecords the generated architecture settings.training_args.jsonrecords the default experiment recipe.model.safetensorsis a valid initialization checkpoint for smoke tests; it is not presented as a trained benchmark checkpoint.- No benchmark score is claimed in this repository.
Architecture
| Item | Value |
|---|---|
| Architecture | Cnn Transformer |
| Scale | giant |
| Attention | flash |
| Fusion | cross attention |
| Activation | relu |
| Normalization | groupnorm |
Default experiment recipe
The included configuration uses novograd with a constant warmup schedule. These are starting values in the script, not evidence of a completed run. For a meaningful evaluation, train all baselines with the same data exposure, tuning budget, and random seeds.
Quick check
python run.py --help
Inspect the script's __main__ block for its generated smoke-test example. Because this is a custom implementation, generic automatic loading APIs require an explicit adapter before use.
Evaluation guidance
A useful first evaluation would use a task-specific labeled split, report the task metric across at least three seeds, and include a matched-capacity baseline. Keep training logs and environment versions with any published result.
Limitations
The initialization checkpoint has not been trained or audited for robustness, fairness, or domain transfer. The implementation should be treated as an experimental starting point. Results from a future trained checkpoint must be documented separately from the defaults shipped here.
Files
run.py— primary artifactREADME.md— this documentationconfig.json— architecture configurationtraining_args.json— default experiment settingsmodel.safetensors— initialization checkpoint
License
Released under bsd-3-clause. Review the source-data terms separately when this repository is used with external datasets.
Configuration
- Architecture
- CustomResearchModel
- Context length (tokens)
- 512
- Layers
- 3
- Hidden size
- 128
- Feed-forward size
- 256
- Attention heads
- 4
- Model type
- cnn_transformer
Identity and Version
- Repository
- Costale90/classification-proto
- Publisher
- Leandro.costa
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 16,576 parameters
- Languages
- Not stated by the source
- Revision
- fa976936e5a4adf36fb5c562fae819392f7bbeff
- First published
- 2026-10-16
- Last updated
- 2026-10-16
Files and Weights
6 files, 76.1 KB in total. The weights are 1 file totalling 66.8 KB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 66.8 KB | 4ed61401de93 |
| config.json | Configuration | 448 B | — |
| run.py | Configuration | 4.6 KB | — |
| training_args.json | Configuration | 200 B | — |
| README.md | Documentation | 2.5 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- bsd-3-clause
- Access
- Open weights, no gate
- Download size
- 66.8 KB
Released by Leandro.costa through its official repository on Hugging Face. Read the license.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 66.8 KB |
| 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 classification-proto
How much GPU memory does classification-proto need?
About 0 GB at 16-bit and 0 GB at 4-bit: the weights (16,576 parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run classification-proto 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 classification-proto commercially?
Yes. classification-proto is released under BSD 3-Clause License. The BSD 3-Clause License is permissive. It permits commercial use and redistribution with the copyright notice, and forbids using the authors' names to endorse derived products without permission.
What is classification-proto's context length?
512 tokens, from the maximum position embeddings in its published configuration.