Bundled offline intelligence components

Qwen3.5-2B GGUF and mmproj-F16.gguf
Source: https://huggingface.co/unsloth/Qwen3.5-2B-GGUF
Upstream model: Qwen/Qwen3.5-2B
License: Apache License 2.0 (see Apache-2.0.txt)

Qwen3-Embedding-0.6B-Q8_0.gguf
Source: https://huggingface.co/Qwen/Qwen3-Embedding-0.6B-GGUF
Upstream model: Qwen/Qwen3-Embedding-0.6B
License: Apache License 2.0 (see Apache-2.0.txt)

Whisper large-v3 turbo Core ML weights
Source: https://huggingface.co/argmaxinc/whisperkit-coreml
Upstream model: OpenAI Whisper
WhisperKit and model license: MIT (see WhisperKit-LICENSE and Whisper-LICENSE)

WeSpeaker speaker-embedding model
Source: https://github.com/wenet-e2e/wespeaker
License: Apache License 2.0 (see Apache-2.0.txt)

llama.cpp runtime
Source: https://github.com/ggml-org/llama.cpp
License: MIT (see llama.cpp-LICENSE)

All model inference runs locally. The application does not upload media to
these upstream projects or services.