Starstruk Logo MarkStarstruk

runtime-llamacpp-v0.1.2

luminal-ai/luminalruntime-llamacpp-v0.1.2Jun 21, 2026by LauraGPT

AI Summary

Prebuilt self-contained binaries for running SenseVoice locally using the llama.cpp/GGUF runtime, featuring built-in FSMN-VAD and strong performance on Chinese and Cantonese.

Key Highlights

  • New: q8 GGUF models are ~half the size of f16 with the same accuracy
  • Prebuilt binaries available for Linux (x64/arm64), macOS (arm64), and Windows (x64)
  • No Python, no build required
  • Built-in FSMN-VAD for on-device ASR

New Features

  • Support for q8 GGUF models
  • Reduced model size without sacrificing accuracy

Full Release Notes

Prebuilt self-contained binaries for running **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.

**New:** q8 GGUF models are ~half the size of f16 with the same accuracy.

```bash
bash download-funasr-model.sh sensevoice ./gguf
llama-funasr-sensevoice -m ./gguf/sensevoice-small-q8.gguf --vad ./gguf/fsmn-vad.gguf -a audio.wav
```

No Python, no build. Linux (x64/arm64), macOS (arm64), Windows (x64). Docs: `runtime/llama.cpp/README.md`.