Open-weight model
tiny-transformer-generation36
by Angel Gunawan angelgag/tiny-transformer-generation36
tiny-transformer-generation36 is an open-weight model from Angel Gunawan, released under MIT License. It has 33,088 parameters and a 128-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.
A research-oriented Tiny Transformer prototype targeting Generation. The included large setup documents defaults and file formats without presenting unverified performance numbers.
Runs On
What it takes to serve tiny-transformer-generation36 (33,088 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.
tiny-transformer-generation36 on every accelerator the SAVRN Index prices, at every precision
Model Card
By Angel Gunawan, published under mit, revision a8038effa65f.
A research-oriented Tiny Transformer prototype targeting Generation. The included large setup documents defaults and file formats without presenting unverified performance numbers. - 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 adam with a exponential schedule. These are starting values in the script, not evidence of a completed…
Read Angel Gunawan's full model card
Tiny Transformer for Generation
Overview
A research-oriented Tiny Transformer prototype targeting Generation. The included large setup documents defaults and file formats without presenting unverified performance numbers.
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 | Tiny Transformer |
| Scale | large |
| Attention | sliding window |
| Fusion | co attention |
| Activation | gelu tanh |
| Normalization | layernorm |
Default experiment recipe
The included configuration uses adam with a exponential 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 held-out set, 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 mit. Review the source-data terms separately when this repository is used with external datasets.
Configuration
- Architecture
- CustomResearchModel
- Context length (tokens)
- 128
- Layers
- 8
- Hidden size
- 256
- Feed-forward size
- 1,024
- Attention heads
- 4
- Model type
- tiny_transformer
Identity and Version
- Repository
- angelgag/tiny-transformer-generation36
- Publisher
- Angel Gunawan
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 33,088 parameters
- Languages
- Not stated by the source
- Revision
- a8038effa65fa02b2d95a91cb722efb93e053870
- First published
- 2026-10-09
- Last updated
- 2026-10-09
Files and Weights
6 files, 142.1 KB in total. The weights are 1 file totalling 132.8 KB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 132.8 KB | e3990cda2590 |
| config.json | Configuration | 456 B | — |
| run.py | Configuration | 4.7 KB | — |
| training_args.json | Configuration | 192 B | — |
| README.md | Documentation | 2.5 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- mit
- Access
- Open weights, no gate
- Download size
- 132.8 KB
Released by Angel Gunawan through its official repository on Hugging Face. Read the license.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 132.8 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 tiny-transformer-generation36
How much GPU memory does tiny-transformer-generation36 need?
About 0 GB at 16-bit and 0 GB at 4-bit: the weights (33,088 parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run tiny-transformer-generation36 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 tiny-transformer-generation36 commercially?
Yes. tiny-transformer-generation36 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.
What is tiny-transformer-generation36's context length?
128 tokens, from the maximum position embeddings in its published configuration.