• Revamped LLM on Genie tutorial: With export script improvements in AI Hub Models, redundant steps were removed from the tutorial.

Fix incorrect file paths in the Whisper demo README file.

Bug fixes:

• Bump Stable Diffusion demo to qai-hub-models 0.37.1 to avoid AIMET-ONNX dependency that blocked demo.

Updates:

• Update QNN TFLite Delegate version to 2.37.0.

• Simplify the process of getting model assets for Whisper and Stable Diffusion.

• "LLM On-Device Deployment" updates

  • Add Falcon3-7B-Instruct.
  • Explain how to properly feed newlines into the Genie CLI.

• Fix typo in LLM UTF-8 Powershell instructions.

• New features
  • Add Llama-SEA-LION-v3.5-8B-R to LLM tutorial
• Bug fixes
  • Windows CLI apps (Classification/SuperResoultion/ObjectDetection)
    • Worked around a vcpkg bug (resolves #42)
    • Worked around issue in QNN EP not copying the necessary files
  • Android apps
    • Fixed links to quantized variants (model cards are now combined and these were dead links)

New Features

• Add Stable Diffusion python demo for Windows.
• Update Whisper python demo for Windows to support streaming from an audio device.
• Add new PowerShell-based python environment setup scripts for Windows demos.

Bug Fixes

• Fix issue where running the Whisper demo with a new version of AI Hub Models would throw an AssertionError.
• Update references to large language models that have been renamed in AI Hub Models.
• Minor bugfixes to the Windows ChatApp.

Minor bugfixes for ChatApp.

Minor updates

• Starting from version 2.32, "QNN SDK" is now referred to as "QAIRT SDK" (with updated download link across several App READMEs).
• Tweaks to LLM tutorial for clarity.
• Add Llama3-TAIDE-LX-8B-Chat-Alpha1 to LLM tutorial.
• Change deprecated soc_id to soc_model in some HTP configurations.

Bug fixes

• Fix bug in deployable ONNX script (tutorials/deploy/onnx/build_deployable_asset.py) affecting numerical correctness.
• Fix demo video (did not work for many users)

Release Notes:

• Added support for IBM Granite 3.1 in LLM Tutorial
• Fixed bug in Whisper demo on Snapdragon X Elite
• Fix poor Super Resolution outputs around black/white regions

What’s new

• Added QCS8550 chipset in Android ChatApp
• Added list of Production devices and OS to run Android ChatApp

Fixes and updates

• Fixed links and documentation for llm-on-genie tutorial
• Update TFLite-QNN Maven Repository name to LiteRT-QNN