---

QAI AppBuilder

QAI AppBuilder (the "Quick AI Application Builder", this repository) is an open-source, on-device AI application platform built on top of the Qualcomm® AI Runtime SDK. Just describe the app you want in plain language — the AI Agent turns it into a complete, runnable application on the Snapdragon NPU, with no coding required.

From an idea to a running app — in one conversation

At the center of QAI AppBuilder is the App Builder: tell the Agent what you want to build ("a screenshot OCR tool", "a voice-memo transcriber", "a real-time speech translator") and it assembles a complete, self-contained local application for you — picking the right models, wiring up the pipeline, and generating the UI. The default output is a lightweight local web app; with a custom prompt the Agent can produce a CLI tool, a desktop utility, or a batch script just as easily.

To do this, the Agent draws on on-device Model Packs (Whisper, Zipformer-ZH, MeloTTS, PP-OCR and more) and, whenever it needs a model that isn't packaged yet, resolves it automatically — downloading a pre-built model straight from Qualcomm AI Hub when one exists, or converting your own PyTorch / ONNX model (export → quantization → context-binary generation → accuracy validation via QAIRT SDK) when it doesn't. Either way, you never leave the conversation, and every model ends up running on the same on-device NPU engine (QNNContext via qai-appbuilder).

Why it matters

• Minutes, not days — get a working local AI app without hand-writing a UI or wiring up model I/O yourself.
• Truly on-device — every model runs on the Snapdragon NPU; your data never leaves the machine. No internet connection is needed at inference time.
• Private by design — a natural fit for privacy-sensitive scenarios (corporate documents, medical images, personal recordings), backed by a built-in security module that gates every file, command, and tool the Agent touches.
• Extensible — the Agent is driven by hot-reloadable Skills and Model Packs; add a new self-contained module and the Agent picks it up automatically.

---

Get Started with QAI AppBuilder

QAI AppBuilder is the natural-language Agent platform for the App Builder, Model Builder, and AI Hub Model Run capabilities described above all live here. Grab the pre-built package below and you'll have a running, on-device AI app builder in two commands.

Just want to use it? Two steps:

Platform: Windows on Snapdragon (ARM64). No admin rights needed — Setup.bat automatically downloads Python, Node.js, QAIRT SDK, and model weights for you.

1. Download & unzip qaiappbuilder.zip, then open cmd.exe in the extracted folder.

2. Run two commands — the first installs the environment (one-time), the second launches the WebUI and opens your browser:

Setup.bat
Start.bat

Your browser opens the QAI AppBuilder WebUI — start chatting to build an app, convert a model, or run one straight from AI Hub.

Want to read or modify the code? The complete QAI AppBuilder project lives at tools/qaiappbuilder. Clone the repo, then run Setup.bat → Build.bat → Start.bat to launch from source.

📖 Docs inside the QAI AppBuilder project:

Document	Description
README | 中文	Full project overview — architecture, features, configuration, skill system, FAQ
Quick Start | 中文	1-page cheat-sheet — dev mode / desktop app (Tauri) / release build, and which .bat to run when

---

What can QAI AppBuilder do?

Supports ARM64 Windows, Linux and Ubuntu (e.g. X Elite Windows, QCS8550 Linux, QCM6490 Ubuntu). Use "native" mode I/O to improve data throughput — see User Guide and Whisper sample for reference.

---

QAI Skills

QAI AppBuilder ships two built-in AI Agent Skills that together cover the full on-device model lifecycle — from sourcing a model all the way to running inference on the Snapdragon NPU.

Skill 1 — AI Hub Model Run

Use this skill when the model you need already exists on Qualcomm AI Hub.

The AI Hub Model Run skill downloads pre-exported QNN models directly from AI Hub and runs them on the Snapdragon NPU via qai_appbuilder (QNNContext) — no conversion, no Visual Studio, no QAIRT SDK required.

Example trigger prompts:

"Download Inception V3 from AI Hub and run inference on my photo"
"Run YOLOv8 object detection on this image using the NPU"
"Use the pre-built Whisper model to transcribe my audio file"

---

Skill 2 — Model Builder

Use this skill when you have a custom PyTorch or ONNX model that is NOT available on AI Hub.

The Model Builder skill automates the full QNN conversion pipeline for custom models — from ONNX export all the way to a validated .bin context binary running on the Snapdragon NPU.

Example trigger prompts:

"Convert my ResNet ONNX model to QNN FP16 and validate the accuracy"
"Convert my custom YOLOv8 model to W8A8 with calibration data and compare against the original"
"Export my PyTorch model to QNN, run inference, and generate an App Builder pack"

---

Quick Start

Python

pip install qai-appbuilder

C++

Download the prebuilt binary package from Releases:

QAI_AppBuilder-win_arm64-{Qualcomm® AI Runtime SDK version}-Release.zip

Refer to User Guide for full API usage, or follow tutorial.ipynb to set up and run a CV model step by step.

Environment Setup

Refer to python.md for instructions on setting up the Python (x64) environment to use QAI AppBuilder on Windows on Snapdragon (WoS) platforms.

You can also run the batch file from QAI AppBuilder Launcher to set up the environment automatically — enabling you to experience the core functionalities within an hour.

---

Diagram

---

WebUI AI Application

We have developed several WebUI AI applications based on QAI AppBuilder, allowing you to experience them quickly. All these applications run on a local PC, requiring no internet connection and are completely free. You can run WebUI AI applications through the batch file 4.Start_WebUI.bat.

Note: Before trying other functions, we suggest that you try these WebUI AI applications first.

---

OpenAI Compatible API Service (GenieAPIService)

Considering that the current mainstream method for invoking LLMs is based on OpenAI-compatible APIs, we have implemented such interfaces in both C++ and Python. This allows application developers to interact with the local large language model running on NPU in a familiar way.

Many third-party applications that support the OpenAI API can seamlessly switch to the local NPU-based model by simply changing the API endpoint.

We have also implemented client sample code for GenieAPIService in both C++ and Python for developer reference.

1. Python based service: Guide to run OpenAI compatible API services developed with Python.
2. C++ based service: Guide to run OpenAI compatible API services developed with C++.

---

Samples

We have a rich set of samples covering multiple categories. All models are sourced from AI Hub and automatically downloaded on first run.

Use the interactive launcher to run any sample without writing code:

cd qai-appbuilder\samples
python run_inference.py              # interactive menu
python run_inference.py --list       # list all available models
python run_inference.py --model whisper_base_en --args "--audio_file input.wav"

See samples/README.md for the full guide including environment setup, model download instructions, and run examples.

---

Tools

1. QAI AppBuilder Launcher

QAI AppBuilder Launcher — enables you to experience the core functionalities of QAI AppBuilder within an hour.

2. DLC2BIN

DLC2BIN — converts the general DLC model format into the BIN format optimized for a specific platform.

3. ONNX2BIN

ONNX2BIN — converts the ONNX model format into the BIN format optimized for a specific platform.

4. ONNXWRAPPER

ONNXWRAPPER — a wrapper to run ONNX inference code with a QNN model, which switches to the QNN runtime automatically.

5. SKILLS

SKILLS includes 3 skills:

• genie_api_service — GenieAPIService technical documentation retrieval
• qai_app_builder — QAI AppBuilder technical documentation retrieval
• qai-runner-skill — QAIRT model conversion & inference on Qualcomm devices

---

Models

Model Hub

LLM Models

---

Blog & Documentation

Official Docs

Blog Posts

The following blog posts are in Chinese (中文).

---

Community Apps

On-device AI apps built by the community on top of QAI AppBuilder — every app runs locally on the Snapdragon NPU, no internet needed at inference time. Browse the full gallery (with category filters and a contributor wall) in CommunityApps/; each app is auto-discovered from its app.json.

Want your app here? See the Community Apps submission guide or post in Discussions → Show & Tell. Add a folder with an app.json, run python CommunityApps/build_gallery.py, and your app appears in the gallery, the index table, and the contributor wall.

---

Third-Party App List

---

QAI AppBuilder Components

There are two ways to use QAI AppBuilder:

1. C++ Library

Download the prebuilt binary package from Releases and link the headers and .lib/.so files into your C++ project:

QAI_AppBuilder-win_arm64-{Qualcomm® AI Runtime SDK version}-Release.zip

2. Python Binding

Install via pip (see Quick Start) or download a specific wheel for your Python version from Releases.

---

Build

You can use the pre-compiled version directly — download the version you need from Releases. To build other wheel variants, refer to BUILD.md.

---

License

QAI AppBuilder is licensed under the BSD 3-clause "New" or "Revised" License. Check out the LICENSE for more details.

---

Star History

---

Disclaimer

This software is provided "as is," without any express or implied warranties. The authors and contributors shall not be held liable for any damages arising from its use. The code may be incomplete or insufficiently tested. Users are solely responsible for evaluating its suitability and assume all associated risks.

Note: Contributions are welcome. Please ensure thorough testing before deploying in critical systems.