GLM-ASR-Nano needs about 3.2 GB of memory at Q8_0 (a 2.4 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 1.3 GB (Q4_K) to 4.5 GB.
Compact 2.26B‑parameter speech recognition model
GLM-ASR-Nano is a 2.26B‑parameter speech recognition model based on the GLM family. It supports Mandarin, Cantonese, and English, and is optimized for low‑volume and noisy audio. The model is released under the MIT license.
Good at
- Low‑volume speech transcription
- Dialect‑aware Mandarin and Cantonese
- Compact size for edge deployment
Limits
- No vision support
- Limited to Mandarin, Cantonese, and English
- Requires 8k context for memory estimates
Best for
- Transcribing quiet meetings
- Deploying on devices with limited VRAM
- Processing long audio in single pass
Who it’s for. Developers building speech‑to‑text applications that need robust performance on Mandarin and Cantonese dialects. The model is suitable for edge devices and low‑resource environments due to its compact size.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K | 1.3 GB | 2.0 GB | Fits | Get file → | |
| Q8_0 | 2.4 GB | 3.2 GB | Fits | Get file → | |
| — | 4.5 GB | 5.6 GB | Fits | Get file → |
For 8 GB VRAM, use the Q4_K variant (1.33 GB, ~2 GB memory at 8k context). If more memory is available, Q8_0 (2.44 GB, ~3.2 GB memory) offers higher precision. The unknown quant is larger (4.52 GB, ~5.6 GB memory) and is recommended only for high‑end systems.
What GPU do you need for GLM-ASR-Nano?
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 Q8_0 (2.4 GB file, about 3.2 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (4.5 GB file, about 5.6 GB in use).
Which GLM-ASR-Nano quant should you choose?
Q8_0 (2.4 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 1.3 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 GLM-ASR-Nano
How much VRAM does GLM-ASR-Nano need?
About 3.2 GB at Q8_0 with an 8K context window. That is the 2.4 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 GLM-ASR-Nano on an 8 GB GPU?
Yes, with the smaller files: the 4.5 GB file needs about 5.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of GLM-ASR-Nano GGUF?
It depends on the quant: Q4_K is 1.3 GB, Q8_0 is 2.4 GB, file is 4.5 GB.
Where can I download GLM-ASR-Nano GGUF?
The files are on Hugging Face in the cstr/glm-asr-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 GLM-ASR-Nano use?
Hugging Face lists the license as mit. 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 GLM-ASR-Nano on 8 GB VRAM?
Yes, the Q4_K variant requires about 2 GB of VRAM at 8k context, fitting comfortably in an 8 GB budget.
Does the model support other languages besides Mandarin and Cantonese?
No, the model is limited to Mandarin, Cantonese, and English; it does not support other languages.
What context length does the model use?
The memory estimates are based on an 8k token context length.