Hush-Nano-22M is an open-weight model from Leecz, released under Apache License 2.0. It has 23M parameters and a 1,024-token context. At 16-bit it needs about 0.1 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
Runs On
What it takes to serve Hush-Nano-22M (23M 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 | 0.1 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 7, 2026.
Hush-Nano-22M on every accelerator the SAVRN Index prices, at every precision
Model Card
The publisher has not written a card for this model.
Configuration
- Architecture
- HushNanoForCausalLM
- Context length (tokens)
- 1,024
- Layers
- 12
- Hidden size
- 384
- Feed-forward size
- 1,024
- Attention heads
- 6
- Key/value heads
- 3
- Vocabulary size
- 8,192
- RoPE base
- 10000
- Stored precision
- float32
- Model type
- hush_nano
Identity and Version
- Repository
- Soulitude/Hush-Nano-22M
- Publisher
- Leecz
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 23M parameters
- Languages
- Not stated by the source
- Revision
- 018c3759a0a22018411c6f4803238bab58efef51
- First published
- 2026-09-24
- Last updated
- 2026-09-24
Files and Weights
9 files, 91.6 MB in total. The weights are 1 file totalling 91.0 MB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 91.0 MB | aa29e283918c |
| config.json | Configuration | 682 B | — |
| configuration_hushnano.py | Configuration | 1.5 KB | — |
| modeling_hushnano.py | Configuration | 13.0 KB | — |
| special_tokens_map.json | Configuration | 107 B | — |
| README.md | Documentation | 28 B | — |
| .gitattributes | Repository | 1.5 KB | — |
| tokenizer.json | Tokenizer | 552.3 KB | — |
| tokenizer_config.json | Tokenizer | 183 B | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 91.0 MB
Released by Leecz through its official repository on Hugging Face. Read the license.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 91.0 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 Hush-Nano-22M
How much GPU memory does Hush-Nano-22M need?
About 0.1 GB at 16-bit and 0 GB at 4-bit: the weights (23M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run Hush-Nano-22M 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 Hush-Nano-22M commercially?
Yes. Hush-Nano-22M 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.
What is Hush-Nano-22M's context length?
1,024 tokens, from the maximum position embeddings in its published configuration.