X-ASR-zh-en streaming zipformer2 needs about 0.6 GB of memory at Q4_K (a 0.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 17 quants, from 0.1 GB (Q3_K) to 0.3 GB (F16).
X-ASR-zh-en streaming zipformer2 transducer for Mandarin–English code‑switching
X-ASR-zh-en is an offline‑streaming ASR model built with the Zipformer2 architecture. It supports multiple chunk sizes (160 ms, 480 ms, 960 ms, 1920 ms) and delivers punctuation and casing. The model is available in GGUF format for rapid deployment with ggml.
Good at
- Low‑latency streaming ASR
- Mandarin–English code‑switching
- Punctuation and casing support
Limits
- No vision support
- Requires GPU or CPU with ggml
- License is apache‑2.0
Best for
- Real‑time transcription of bilingual audio
- Offline voice‑to‑text on edge devices
- Deploying ASR in low‑latency applications
Who it’s for. Engineers and researchers building speech recognition systems that need a compact, low‑latency model for Mandarin–English code‑switching. The model is suitable for developers who prefer GGUF and ggml backends.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q3_K | 0.1 GB | 0.6 GB | Fits | Get file → | |
| Q3_K | 0.1 GB | 0.6 GB | Fits | Get file → | |
| Q3_K | 0.1 GB | 0.6 GB | Fits | Get file → | |
| Q3_K | 0.1 GB | 0.6 GB | Fits | Get file → | |
| IQ4_XS | 0.1 GB | 0.6 GB | Fits | Get file → | |
| Q4_K | 0.1 GB | 0.6 GB | Fits | Get file → | |
| Q4_K | 0.1 GB | 0.6 GB | Fits | Get file → | |
| Q4_K | 0.1 GB | 0.6 GB | Fits | Get file → | |
| Q4_K | 0.1 GB | 0.6 GB | Fits | Get file → | |
| Q8_0 | 0.2 GB | 0.7 GB | Fits | Get file → | |
| Q8_0 | 0.2 GB | 0.7 GB | Fits | Get file → | |
| Q8_0 | 0.2 GB | 0.7 GB | Fits | Get file → | |
| Q8_0 | 0.2 GB | 0.7 GB | Fits | Get file → | |
| F16 | 0.3 GB | 0.8 GB | Fits | Get file → | |
| F16 | 0.3 GB | 0.8 GB | Fits | Get file → | |
| F16 | 0.3 GB | 0.8 GB | Fits | Get file → | |
| F16 | 0.3 GB | 0.8 GB | Fits | Get file → |
For 8 GB VRAM, the Q4_K (0.09 GB) or IQ4_XS (0.09 GB) files are suitable, estimating 0.6 GB memory at 8 k context. For 4 GB VRAM, use Q3_K (0.07 GB) files, also estimating 0.6 GB memory. For lossless quality, choose Q8_0 (0.16 GB) or F16 (0.31 GB) if memory allows.
What GPU do you need for X-ASR-zh-en streaming zipformer2?
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 (0.3 GB file, about 0.8 GB in use).
Which X-ASR-zh-en streaming zipformer2 quant should you choose?
Q4_K (0.1 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.
If memory is tight, the smallest file is Q3_K at 0.1 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about X-ASR-zh-en streaming zipformer2
How much VRAM does X-ASR-zh-en streaming zipformer2 need?
About 0.6 GB at Q4_K with an 8K context window. That is the 0.1 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 X-ASR-zh-en streaming zipformer2 on an 8 GB GPU?
Yes, with F16: the 0.3 GB file needs about 0.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of X-ASR-zh-en streaming zipformer2 GGUF?
It depends on the quant: Q3_K is 0.1 GB, Q3_K is 0.1 GB, Q3_K is 0.1 GB, Q3_K is 0.1 GB, IQ4_XS is 0.1 GB, Q4_K is 0.1 GB and 11 more.
Where can I download X-ASR-zh-en streaming zipformer2 GGUF?
The files are on Hugging Face in the Luigi/x-asr-zh-en-streaming-zipformer2-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 X-ASR-zh-en streaming zipformer2 use?
Hugging Face lists the license as apache-2.0. 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 X-ASR-zh-en on 4 GB VRAM?
Yes, the Q3_K quantized files are 0.07 GB and estimate 0.6 GB memory at 8 k context, which fits 4 GB VRAM.
Does the model support vision tasks?
No, X-ASR-zh-en is an ASR model and does not include vision capabilities.
What is the license of this model?
The model is released under the apache‑2.0 license.