v0.30.1

qai-hub-apps CLI just got a major usability upgrade

We’re excited to announce a major upgrade to the qai-hub-apps developer experience
Getting started with AI Hub Apps is now simpler, faster, and far more reproducible.

What's New

You can now fetch and set up apps with a single command:

qai-hub-apps fetch <app_id> --model <model_id> --chipset <chipset>

We now streamline the entire setup experience by:

• Automatically downloading supported models
• Providing helper scripts for setup/build/run workflows
• Including Docker-based workflows for reproducible environments
• Shipping updated READMEs with simplified instructions

Get started

pip install qai-hub-apps

PyPi: https://pypi.org/project/qai-hub-apps/
GitHub: https://github.com/qualcomm/ai-hub-apps

Additional Notes:

• Updated the GitHub branch strategy:
  • main is now the active development branch
  • release now contain validated/released app sources
• Removed Llama 2 7B from chatapp_windows

v0.29.0

Improvements

• Updated LLM tutorial (make it clear assets can be downloaded; general updates)

v0.28.0

Releasing AI Hub Apps CLI which can be used to fetch apps with supported models from AI Hub Models

v0.26.0

• Android ChatApp shows TPS/TTFT measurements in the UI
• Object Detection Android app added to website
• Fixed Stable Diffusion demo to ensure correct scheduler is instantiated

v0.24.0

• Renamed app status terminology from public/private to published/unpublished

v0.23.0

• Added MediaPipe Hand Gesture Recognition

v0.22.0

• Minor updates to LLM tutorial
• Apps now use QAIRT 2.42

v0.20.0

Updates:

• QNN version bumped to 2.40 for Android apps (not including ChatApp)
• Improved Android ChatApp (removes flicker and improves scrolling)

v0.19.0

• The build for each Android application is now completely independent of other apps. Each app should be opened in Android Studio in the individual app's folder, rather than opening the parent Android folder.
• Added support in ChatApp for Snapdragon 7 Gen 4

v0.18.0

• Added instructions on how to increase swap space for LLM export in case of out-of-memory errors.
• Updated import paths to work with Qualcomm AI Hub Model v0.41.0