Fun-ASR-Nano-2512 needs about 1.9 GB of memory at Q8_0 (a 1.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 0.9 GB (Q4_K) to 2.0 GB (F16).
Fun-ASR-Nano-2512: 0.985B parameter speech‑recognition model in GGUF format
Fun-ASR-Nano-2512 is a 0.985 B parameter end‑to‑end ASR model. It supports 31 languages and is optimized for far‑field, high‑noise environments. The model is available in GGUF format for llama.cpp inference.
Good at
- Far‑field high‑noise recognition
- Multi‑language support
- Low‑latency real‑time transcription
Limits
- No vision support
- No CTC path, only LLM‑decoder
- License requires attribution
Best for
- Real‑time transcription on edge devices
- Industry‑specific terminology recognition
- Low‑latency voice interfaces
Who it’s for. Developers building voice‑enabled applications that need accurate transcription across multiple languages. The model is suitable for edge deployments using llama.cpp or CrispASR runtimes.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K | 0.9 GB | 1.5 GB | Fits | Get file → | |
| Q8_0 | 1.3 GB | 1.9 GB | Fits | Get file → | |
| F16 | 2.0 GB | 2.7 GB | Fits | Get file → |
For 8 GB VRAM, use Q4_K (0.9 GB, ~1.5 GB memory at 8k context). For 16 GB VRAM, Q8_0 (1.27 GB, ~1.9 GB memory). For 32 GB VRAM, F16 (1.98 GB, ~2.7 GB memory). Estimates are approximate.
What GPU do you need for Fun-ASR-Nano-2512?
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 (2.0 GB file, about 2.7 GB in use).
Which Fun-ASR-Nano-2512 quant should you choose?
Q8_0 (1.3 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
If memory is tight, the smallest file is Q4_K at 0.9 GB, with the quality trade-offs that come with fewer bits. 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 Fun-ASR-Nano-2512
How much VRAM does Fun-ASR-Nano-2512 need?
About 1.9 GB at Q8_0 with an 8K context window. That is the 1.3 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 Fun-ASR-Nano-2512 on an 8 GB GPU?
Yes, with F16: the 2.0 GB file needs about 2.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Fun-ASR-Nano-2512 GGUF?
It depends on the quant: Q4_K is 0.9 GB, Q8_0 is 1.3 GB, F16 is 2.0 GB.
Where can I download Fun-ASR-Nano-2512 GGUF?
The files are on Hugging Face in the cstr/funasr-nano-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 Fun-ASR-Nano-2512 use?
Hugging Face lists the license as other. 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 Fun-ASR-Nano-2512 on 8 GB VRAM?
Yes, the Q4_K quantized file is 0.9 GB and requires about 1.5 GB of VRAM at 8k context.
Does the model support vision tasks?
No, the model only performs speech recognition and has no vision capabilities.
What license applies to the GGUF files?
The GGUF files are released under the FunASR Model License v1.1, which allows commercial use with attribution.