Automatic Speech Recognition
GGUF
Chinese
GGUF
confucius4
r2t2
asr
streaming
real-time
low-latency
speech-recognition
multilingual
conversational
How to use from
llama.cppInstall from WinGet (Windows)
winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf netease-youdao/Confucius4-R2T2-GGUF:# Run inference directly in the terminal:
llama cli -hf netease-youdao/Confucius4-R2T2-GGUF:Use pre-built binary
# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf netease-youdao/Confucius4-R2T2-GGUF:# Run inference directly in the terminal:
./llama-cli -hf netease-youdao/Confucius4-R2T2-GGUF:Build from source code
git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf netease-youdao/Confucius4-R2T2-GGUF:# Run inference directly in the terminal:
./build/bin/llama-cli -hf netease-youdao/Confucius4-R2T2-GGUF:Use Docker
docker model run hf.co/netease-youdao/Confucius4-R2T2-GGUF:Quick Links
Confucius4-R2T2: A Low Latency and High Accuracy Real-Time Speech Recognition Model
Real Real-Time Transcription
Confucius4-R2T2-GGUF
Official GGUF release from NetEase Youdao for netease-youdao/Confucius4-R2T2, a low-latency and high-accuracy true streaming Automatic Speech Recognition (ASR) model that features fine-grained and configurable decoding chunks from 80 ms to 2 s.
Files
| Filename | Quantization | Size | Notes |
|---|---|---|---|
| Confucius4-R2T2-f16.gguf | f16 | 3.2GiB | best quality |
| Confucius4-R2T2-Q8_0.gguf | Q8_0 | 1.7GiB | very good quality |
| Confucius4-R2T2-Q4_K_M.gguf | Q4_K_M | 1.0GiB | fast, lower quality |
| mmproj-Confucius4-R2T2-f16.gguf | mmproj-f16 | 0.6GiB | multi-modal supplement |
| mmproj-Confucius4-R2T2-Q8_0.gguf | mmproj-Q8_0 | 0.3GiB | multi-modal supplement |
The low-bit variants are quantized from the f16 GGUF.
- Downloads last month
- 5,518
Hardware compatibility
Log In to add your hardware
4-bit
8-bit
16-bit
Install (macOS, Linux)
# Start a local OpenAI-compatible server with a web UI: llama serve -hf netease-youdao/Confucius4-R2T2-GGUF:# Run inference directly in the terminal: llama cli -hf netease-youdao/Confucius4-R2T2-GGUF: