moonshine-tiny needs about 0.5 GB of memory at Q4_K (a 0.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 0.0 GB (Q4_K) to 0.1 GB (F16).
Moonshine Tiny German ASR model in GGUF format
Moonshine Tiny German is a 27M parameter speech recognition model fine‑tuned from English moonshine-tiny. It delivers 11.4% WER and 4.2% CER on Common Voice 22 German test set. The model is licensed under CC‑BY‑NC‑SA‑4.0.
Good at
- German speech recognition
- Low‑latency inference on CPU
- Small model footprint
Limits
- No vision support
- Requires tokenizer.bin in same folder
- License restricts commercial use
Best for
- Transcribing German audio clips
- Generating captions for German videos
- Voice‑to‑text in resource‑constrained devices
Who it’s for. Developers building German ASR applications who need a lightweight model that runs on CPU. The model is suitable for prototyping and edge deployments where GPU resources are unavailable.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K | 0.0 GB | 0.5 GB | Fits | Get file → | |
| F16 | 0.1 GB | 0.6 GB | Fits | Get file → |
For 8 GB VRAM, the Q4_K variant (0.02 GB) is ideal, using about 0.5 GB at 8 k context. If more precision is needed and 8 GB VRAM is available, the F16 variant (0.05 GB) uses about 0.6 GB at 8 k context. Both estimates are approximate.
What GPU do you need for moonshine-tiny?
These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is F16 (0.1 GB file, about 0.6 GB in use).
Which moonshine-tiny quant should you choose?
Q4_K (0.0 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about moonshine-tiny
How much VRAM does moonshine-tiny need?
About 0.5 GB at Q4_K with an 8K context window. That is the 0.0 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run moonshine-tiny on an 8 GB GPU?
Yes, with F16: the 0.1 GB file needs about 0.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of moonshine-tiny GGUF?
It depends on the quant: Q4_K is 0.0 GB, F16 is 0.1 GB.
Where can I download moonshine-tiny GGUF?
The files are on Hugging Face in the cstr/moonshine-tiny-de-fidoriel-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does moonshine-tiny use?
Hugging Face lists the license as cc-by-nc-sa-4.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Moonshine Tiny on 8 GB VRAM?
Yes. The Q4_K quantization uses about 0.5 GB at 8 k context, well within 8 GB VRAM limits.
Does the model support other languages?
No. It is fine‑tuned for German only.
Is GPU required for inference?
No. The model runs on CPU; GPU is not needed for these small models.