CO

Confucius4 R2T2

ConfuciusGGUF by netease-youdaoUpdated 2026-09-20
Download on Hugging Face
Parameters1.72B
Context64K
Layers28
Architectureqwen3vl
Smallest quant1.1 GB
Quick answer

Confucius4 R2T2 needs about 2.5 GB of memory at Q4_K_M (a 1.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 1.1 GB (Q4_K_M) to 3.4 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 3.8 GB.

Confucius4 R2T2 – low‑latency, high‑accuracy real‑time ASR

Confucius4 R2T2 is a true streaming Automatic Speech Recognition model built on Qwen3‑VL. It offers low latency from 80 ms to 2 s, stable output, and maintains offline accuracy. The model supports Chinese and English and can be run with llama.cpp or vLLM.

Good at

  • Low‑latency real‑time transcription
  • Stable, non‑revise output
  • Configurable chunking for latency/accuracy trade‑offs

Limits

  • No vision support beyond basic multimodal supplement
  • Requires 8 k context for memory estimates
  • License type is other, not specified

Best for

  • Real‑time live captioning
  • Downstream NLP pipelines
  • Simultaneous speech translation

Who it’s for. Developers building real‑time transcription services or downstream NLP agents that need instant, non‑revise text. Suitable for applications requiring low latency and high accuracy in Chinese and English.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_M1.1 GB 2.5 GB
Fits Get file →
Q8_01.8 GB 3.3 GB
Fits Get file →
F163.4 GB 4.9 GB
Fits Get file →

For 8 GB VRAM, Q4_K_M (1.11 GB, ~2.5 GB RAM) is recommended. With 12 GB VRAM, Q8_0 (1.83 GB, ~3.3 GB RAM) offers better quality. If higher fidelity is needed and 16 GB VRAM is available, F16 (3.45 GB, ~4.9 GB RAM) can be used.

What GPU do you need for Confucius4 R2T2?

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: the best quant that fits is Q8_0 (1.8 GB file, about 3.3 GB in use).
  • 6 GB VRAM: the best quant that fits is F16 (3.4 GB file, about 4.9 GB in use).

Which Confucius4 R2T2 quant should you choose?

Q4_K_M is the usual starting point for Confucius4 R2T2: it balances file size and quality well. Here it is a 1.1 GB file.

The Q8_0 file is 1.8 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

Architectureqwen3vl
Parameters1.72B
Layers28
Trained context65,536 tokens
KV cache (fp16)112 KB per token

Source

Quantized bynetease-youdao
FamilyConfucius
Licenseother
Downloads (HF)8,016
Last updated2026-09-20

Frequently asked questions about Confucius4 R2T2

How much VRAM does Confucius4 R2T2 need?

About 2.5 GB at Q4_K_M with an 8K context window. That is the 1.1 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 Confucius4 R2T2 on an 8 GB GPU?

Yes, with F16: the 3.4 GB file needs about 4.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Confucius4 R2T2 GGUF?

It depends on the quant: Q4_K_M is 1.1 GB, Q8_0 is 1.8 GB, F16 is 3.4 GB.

What context length does Confucius4 R2T2 support?

The GGUF header lists a trained context length of 65,536 tokens (64K). The KV cache grows with the context you use: about 0.9 GB at 8K and 3.8 GB at 32K.

Where can I download Confucius4 R2T2 GGUF?

The files are on Hugging Face in the netease-youdao/Confucius4-R2T2-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 Confucius4 R2T2 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 Confucius4 R2T2 on 8 GB VRAM?

Yes, the Q4_K_M quantization is 1.11 GB and requires about 2.5 GB of RAM at 8 k context, fitting within 8 GB VRAM.

Does the model support vision tasks?

No, the base model has no vision capability; only a multimodal supplement is provided.

What context length does it support?

It was trained with 65,536 tokens and can operate with 8 k context for inference.

Other GGUFs of Confucius4-R2T2

Similar models

Browse all →

More from netease-youdao

All by netease-youdao →

More Confucius models

All Confucius →

Other tiny models

All tiny models →