runtime-llamacpp-v0.1.1

PaddlePaddle/Paddleruntime-llamacpp-v0.1.1Jun 21, 2026by LauraGPT

AI Summary

A runtime release providing prebuilt, self-contained binaries to run SenseVoice and Paraformer locally via llama.cpp, featuring on-device ASR capabilities and built-in VAD without requiring Python or build tools.

Key Highlights

  • Prebuilt binaries for Linux, macOS, and Windows
  • Local execution of SenseVoice and Paraformer
  • Built-in FSMN-VAD for audio activity detection
  • Improved SenseVoice CER (8.01%) compared to whisper.cpp

New Features

  • SenseVoice local ASR support
  • Paraformer and Fun-ASR-Nano support
  • Built-in FSMN-VAD
  • On-device whisper.cpp-style ASR

Full Release Notes

Prebuilt, self-contained binaries to run **SenseVoice** (and Paraformer / Fun-ASR-Nano) locally with the FunASR llama.cpp / GGUF runtime — built-in FSMN-VAD, whisper.cpp-style on-device ASR, strong on Chinese & Cantonese.

```bash
bash download-funasr-model.sh sensevoice ./gguf
llama-funasr-sensevoice -m ./gguf/SenseVoiceSmall-f16.gguf --vad ./gguf/fsmn-vad.gguf -a audio.wav
```

No Python, no build, no torch. Binaries for Linux (x64/arm64), macOS (arm64), Windows (x64). Docs: `runtime/llama.cpp/README.md`. CER (micro-avg): SenseVoice 8.01% vs whisper.cpp small 22%.