runtime-llamacpp-v0.1.4

modelscope/FunASRruntime-llamacpp-v0.1.4Jun 29, 2026by github-actions[bot]

AI Summary

Prebuilt self-contained binaries for the FunASR llama.cpp runtime, supporting SenseVoice, Paraformer, and Fun-ASR-Nano with built-in FSMN-VAD. Designed for on-device ASR with strong Chinese performance.

Key Highlights

  • Prebuilt self-contained binaries for llama.cpp / GGUF runtime
  • Supports SenseVoice, Paraformer, and Fun-ASR-Nano models
  • Built-in FSMN-VAD for speech segmentation without Python

New Features

  • Self-contained binaries (no Python, no build required)
  • Support for x64 and x64-avx2 CPU assets
  • CLI tools: llama-funasr-cli, llama-funasr-sensevoice, llama-funasr-paraformer

Full Release Notes

Prebuilt self-contained binaries for the FunASR llama.cpp / GGUF runtime — SenseVoice, Paraformer and Fun-ASR-Nano with built-in FSMN-VAD (a whisper.cpp-style on-device ASR, strong on Chinese). Get a model with `bash download-funasr-model.sh <sensevoice|paraformer|nano>`, then run `llama-funasr-cli` / `llama-funasr-sensevoice` / `llama-funasr-paraformer`. Use the default x64 asset for maximum CPU compatibility; use the x64-avx2 asset on CPUs with AVX2/FMA/F16C/BMI2 for higher throughput. No Python, no build. Docs: runtime/llama.cpp/README.md