Hy-SimulMT-30B-A3B needs about 21.1 GB of memory at Q4_K_M (a 19.7 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 2 quants, from 19.7 GB (Q4_K_M) to 33.3 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 3.2 GB.
Hy-SimulMT-30B-A3B: Real‑time simultaneous Chinese‑English translation
Hy‑SimulMT-30B-A3B is a 30.4B parameter model designed for real‑time speech‑to‑speech translation between Chinese and English. It supports continuous audio input, producing translated text and speech with voice cloning. The model is built on the hy_v3 architecture and is a mixture of experts with 3B active parameters.
Good at
- Real‑time simultaneous translation
- Bidirectional speech‑to‑speech
- Voice cloning
Limits
- No vision support
- Requires large VRAM (≈21GB for Q4_K_M)
- Licensed under Apache‑2.0
Best for
- Live meetings
- Speech and interview translation
- Real‑time communication
Who it’s for. Developers building live translation tools or services that need low‑latency, high‑quality Chinese‑English speech translation. The model is suitable for applications requiring continuous audio processing and speaker‑specific voice cloning.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 19.7 GB | 21.1 GB | Too big | Get file → | |
| Q8_0 | 33.3 GB | 34.6 GB | Too big | Get file → |
For systems with 32GB VRAM, the Q4_K_M quantization (19.75GB file, 21.1GB memory at 8k context) is recommended. If more VRAM is available, the Q8_0 quant (33.31GB file, 34.6GB memory) provides higher precision. Memory estimates are approximate.
What GPU do you need for Hy-SimulMT-30B-A3B?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) 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: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: the best quant that fits is Q4_K_M (19.7 GB file, about 21.1 GB in use).
- 32 GB VRAM: the best quant that fits is Q4_K_M (19.7 GB file, about 21.1 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (33.3 GB file, about 34.6 GB in use).
Which Hy-SimulMT-30B-A3B quant should you choose?
Q4_K_M is the usual starting point for Hy-SimulMT-30B-A3B: it balances file size and quality well. Here it is a 19.7 GB file.
The Q8_0 file is 33.3 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Hy-SimulMT-30B-A3B
How much VRAM does Hy-SimulMT-30B-A3B need?
About 21.1 GB at Q4_K_M with an 8K context window. That is the 19.7 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 Hy-SimulMT-30B-A3B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 21.1 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Hy-SimulMT-30B-A3B GGUF?
It depends on the quant: Q4_K_M is 19.7 GB, Q8_0 is 33.3 GB.
What context length does Hy-SimulMT-30B-A3B support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.8 GB at 8K and 3.2 GB at 32K.
Where can I download Hy-SimulMT-30B-A3B GGUF?
The files are on Hugging Face in the tencent/Hy-SimulMT-30B-A3B-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 Hy-SimulMT-30B-A3B 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 Hy‑SimulMT-30B-A3B on 8GB VRAM?
No. The smallest quantization requires about 21GB of VRAM at 8k context.
Does this model support vision tasks?
No. It is a text‑and‑speech model with no vision capabilities.
What license does the model use?
Apache‑2.0.