Extend QVAC with plugins

Discover plugins for QVAC. Enable and disable built-in AI capabilities, and add new ones via custom plugins.

llamacpp-completion

Stable

Text generation via llama.cpp - streaming, multimodal, RAG, and fine-tuning. Runs on CPU/GPU/Metal/CUDA.

llamacpp-embedding

Stable

Text embeddings via llama.cpp for semantic search and RAG pipelines - fully on-device.

nmtcpp-translation

Stable

Offline neural machine translation via nmt.cpp. 200+ language pairs using NLLB-200 models.

onnx-ocr

Stable

On-device optical character recognition via ONNX Runtime. Extracts text from images and documents.

onnx-tts

Stable

On-device text-to-speech via ONNX Runtime. Natural-sounding speech with streaming audio output.

parakeet-transcription

Stable

Speech-to-text via NVIDIA Parakeet. Fast, accurate transcription optimized for on-device inference.

sdcpp-generation

Stable

On-device image generation via sd.cpp (Stable Diffusion). Supports SD 1.5 and SDXL, Metal and CUDA accelerated.

whispercpp-transcription

Stable

Speech-to-text via whisper.cpp. Streaming segments and word-level timestamps, fully on-device.

Community

Built a plugin?

Share it with the community, or learn how to wrap any model into a custom QVAC plugin.