# DeGirum Docs

## DeGirum

- [DeGirum](https://docs.degirum.com/readme.md): DeGirum offers quickstart guides for AI Hub and PySDK, plus resources for fast, efficient edge AI development.

## AI Hub

- [Overview](https://docs.degirum.com/ai-hub/readme.md): Explore the DeGirum AI Hub—your platform for evaluating models, running cloud inference, and compiling models for deployment with your hardware.
- [Quickstart](https://docs.degirum.com/ai-hub/quickstart.md): Learn to sign up, navigate AI Hub, and execute real-time inferencing to quickly bring your ideas to life.
- [Hardware Explorer](https://docs.degirum.com/ai-hub/hardware-explorer.md): Access a diverse range of AI hardware via DeGirum’s Device Farm on AI Hub. Prototype and experiment with cutting-edge accelerators without owning physical devices.
- [Inference Console](https://docs.degirum.com/ai-hub/hardware-explorer/inference-console.md): Explore model and hardware combinations with the AI Hub Inference Console.
- [Public Models](https://docs.degirum.com/ai-hub/public-models.md): Explore the public Model Zoo, a comprehensive collection of pre-optimized AI models for diverse applications.
- [Hailo](https://docs.degirum.com/ai-hub/public-models/hailo.md): This page features models for Hailo devices currently available on the DeGirum AI Hub model zoo.
- [DEEPX](https://docs.degirum.com/ai-hub/public-models/deepx.md): This page features models for DEEPX devices currently available on the DeGirum AI Hub model zoo.
- [Axelera AI](https://docs.degirum.com/ai-hub/public-models/axelera-ai.md): This page features models for Axelera AI devices currently available on the DeGirum AI Hub model zoo.
- [Intel](https://docs.degirum.com/ai-hub/public-models/intel.md): This page features models for Intel devices (OpenVINO multidevice) currently available on the DeGirum AI Hub model zoo.
- [Google](https://docs.degirum.com/ai-hub/public-models/google.md): This page features models for Google devices currently available on the DeGirum AI Hub model zoo.
- [MemryX](https://docs.degirum.com/ai-hub/public-models/memryx.md): This page features models for MemryX devices currently available on the DeGirum AI Hub model zoo.
- [BrainChip](https://docs.degirum.com/ai-hub/public-models/brainchip.md): This page features models for BrainChip devices currently available on the DeGirum AI Hub model zoo.
- [Rockchip](https://docs.degirum.com/ai-hub/public-models/rockchip.md): This page features models for Rockchip devices currently available on the DeGirum AI Hub model zoo.
- [DeGirum](https://docs.degirum.com/ai-hub/public-models/degirum.md): This page features models for DeGirum devices currently available on the DeGirum AI Hub model zoo.
- [Model Console](https://docs.degirum.com/ai-hub/model-console.md): Leverage DeGirum’s Model Console to run models directly in your web browser – upload inputs, view real-time results, and explore model details with ease.
- [Workspace Plans](https://docs.degirum.com/ai-hub/workspace-plans.md): Learn what each workspace plan includes and how to upgrade, add users and devices, and manage billing and usage in AI Hub.
- [Workspaces](https://docs.degirum.com/ai-hub/workspaces.md): Read this page to learn about workspaces in the AI Hub. You must join a workspace to access unique AI Hub features.
- [Cloud Compiler](https://docs.degirum.com/ai-hub/workspaces/cloud-compiler.md): Port and optimize your custom AI models for various hardware platforms using DeGirum’s Cloud Compiler.
- [Workspace Models](https://docs.degirum.com/ai-hub/workspaces/workspace-models.md): Read this page to learn about how to navigate and create model zoos in the DeGirum AI Hub. All users have access to the model zoo. The ability to create model zoos is on a per-workspace basis.
- [Workspace Tokens](https://docs.degirum.com/ai-hub/workspaces/workspace-tokens.md): The AI Hub features Workspace tokens for accessing the AI Hub API. You can create Workspace tokens in the AI Hub as soon as you create an account.
- [Workspace Settings](https://docs.degirum.com/ai-hub/workspaces/workspace-settings.md): Learn about Workspace Settings. Change member roles, invite members, remove members, manage billing, and delete the Workspace.
- [PySDK Integration](https://docs.degirum.com/ai-hub/pysdk-integration.md): Integrate DeGirum PySDK with AI Hub to keep your code concise while tapping into powerful cloud-hosted inference.
- [Application Package Licensing](https://docs.degirum.com/ai-hub/application-package-licensing.md): Understand how Application Package licensing works in DeGirum AI Hub, including automatic license fetch, Workspace eligibility, and billing based on peak active usage.

## PySDK

- [Overview](https://docs.degirum.com/pysdk/readme.md): Discover the power and flexibility of DeGirum PySDK, engineered to streamline AI development and deliver consistent performance across diverse hardware platforms.
- [Quickstart](https://docs.degirum.com/pysdk/quickstart.md): See a live example of PySDK, demonstrating live inferencing and rapid deployment.
- [Installation](https://docs.degirum.com/pysdk/installation.md): Follow comprehensive installation guides for PySDK, covering virtual environments, Docker images, and verification.
- [Token Installation and Management](https://docs.degirum.com/pysdk/token-installation-and-management.md): Create, install, and manage DeGirum Workspace tokens for PySDK, including the one-token-per-device rule, auto-renewal behavior, and secure practices for small setups and large fleets.
- [Runtimes and Drivers](https://docs.degirum.com/pysdk/runtimes-and-drivers.md): This page provides an overview of the runtimes and drivers supported by PySDK.
- [Hailo](https://docs.degirum.com/pysdk/runtimes-and-drivers/hailort.md): This page details the installation procedures for the Hailo runtime in PySDK. It outlines the steps for installing on Raspberry Pi and Ubuntu.
- [OpenVINO](https://docs.degirum.com/pysdk/runtimes-and-drivers/openvino.md): This page details the installation procedures for the OpenVINO Runtime supported by PySDK.
- [Axelera AI](https://docs.degirum.com/pysdk/runtimes-and-drivers/axelera.md): This page details installation procedures for the Axelera runtime in PySDK.
- [MemryX](https://docs.degirum.com/pysdk/runtimes-and-drivers/memryxrt.md): This page provides information on installing the MemryX Runtime for MemryX AI accelerators.
- [BrainChip](https://docs.degirum.com/pysdk/runtimes-and-drivers/akida.md): This page provides step-by-step instructions for installing the Akida (BrainChip) runtime in PySDK.
- [Rockchip](https://docs.degirum.com/pysdk/runtimes-and-drivers/rknn.md): This page details the installation procedures for Rockchip devices.
- [ONNX](https://docs.degirum.com/pysdk/runtimes-and-drivers/onnx-runtime.md): This page provides step-by-step instructions for installing the ONNX runtime in PySDK.
- [TensorRT](https://docs.degirum.com/pysdk/runtimes-and-drivers/tensorrt.md): PySDK supports the NVIDIA TensorRT runtime on Linux, Windows, and NVIDIA Jetson hardware. This page walks through installation on each platform.
- [PySDK User Guide](https://docs.degirum.com/pysdk/user-guide-pysdk.md): Explore DeGirum's PySDK in depth. This guide covers the core concepts.
- [Core Concepts](https://docs.degirum.com/pysdk/user-guide-pysdk/core-concepts.md): Explore the core components of PySDK—including the AI inference engine, AI model, and model zoo—to understand how they power modern edge AI applications.
- [Organizing Models](https://docs.degirum.com/pysdk/user-guide-pysdk/organizing-models.md): Learn how AI Hub models and model zoos are organized. This page covers JSON model file naming conventions and model zoo directory structure. Local model zoos may follow these conventions.
- [Setting Up an AI Server](https://docs.degirum.com/pysdk/user-guide-pysdk/setting-up-an-ai-server.md): Read this page if you'll host an AI server or perform inference with a local server.
- [Loading an AI Model](https://docs.degirum.com/pysdk/user-guide-pysdk/loading-an-ai-model.md): This is an end-to-end guide for loading a model. You'll start with connecting to an inference engine and model zoo, learn about filtering model lists, then loading a model.
- [Running AI Model Inference](https://docs.degirum.com/pysdk/user-guide-pysdk/running-ai-model-inference.md): This is a walkthrough for running predictions. You'll learn about input data types, understanding the results, and finally processing inputs in batches for efficiency.
- [Model JSON Structure](https://docs.degirum.com/pysdk/user-guide-pysdk/model-json-structure.md): This page outlines model JSON structure and parameters.
- [Command Line Interface](https://docs.degirum.com/pysdk/user-guide-pysdk/command-line-interface.md): Learn how to use the PySDK command line interface to manage AI models, control your AI server, and streamline model downloads.
- [API Reference Guide](https://docs.degirum.com/pysdk/user-guide-pysdk/api-ref.md): This page serves as an index for the PySDK API Reference Guide, providing access to detailed documentation of the classes, methods, and properties available in PySDK.
- [PySDK Package](https://docs.degirum.com/pysdk/user-guide-pysdk/api-ref/package.md): PySDK API Reference Guide. PySDK entry points: connect(), LOCAL/CLOUD designators.
- [Model Module](https://docs.degirum.com/pysdk/user-guide-pysdk/api-ref/model.md): PySDK API Reference Guide. Core Model class.
- [Zoo Manager Module](https://docs.degirum.com/pysdk/user-guide-pysdk/api-ref/zoo-manager.md): PySDK API Reference Guide. Load, list and authenticate against local, server or cloud zoos.
- [Postprocessor Module](https://docs.degirum.com/pysdk/user-guide-pysdk/api-ref/postprocessor.md): PySDK API Reference Guide. InferenceResults containers.
- [AI Server Module](https://docs.degirum.com/pysdk/user-guide-pysdk/api-ref/server.md): PySDK API Reference Guide. CLI launcher for the DeGirum AI server.
- [Miscellaneous Modules](https://docs.degirum.com/pysdk/user-guide-pysdk/api-ref/misc.md): PySDK API Reference Guide. Console logging, verbosity control and helper exceptions.
- [Older PySDK User Guides](https://docs.degirum.com/pysdk/user-guide-pysdk/archive.md): Index of older PySDK user guides.
- [Release Notes](https://docs.degirum.com/pysdk/release-notes.md): This page features release notes for releases of PySDK. You may download PySDK versions listed here from PyPI.org.
- [EULA v1.0 and later](https://docs.degirum.com/pysdk/eula.md): Review the DeGirum PySDK End-User License Agreement (EULA) for PySDK v1.0+ to understand the terms governing your use of DeGirum PySDK and any Application Packages designed to operate with it.
- [EULA v0.20.0 and earlier](https://docs.degirum.com/pysdk/eula-v0.20.0-and-earlier.md): Review the DeGirum PySDK End-User License Agreement for PySDK v0.20.0 and earlier to understand the terms governing your use of DeGirum PySDK.

## DeGirum Tools

- [Overview](https://docs.degirum.com/degirum-tools/readme.md): We provide the DeGirum Tools Python package to aid development of AI applications with PySDK. In this group, we'll outline main concepts of DeGirum Tools and provide the API Reference Guide.
- [Model Registry](https://docs.degirum.com/degirum-tools/model_registry.md): DeGirum Tools API Reference Guide. YAML registry for selecting models by task, hardware, and runtime defaults.
- [Inference Support](https://docs.degirum.com/degirum-tools/inference_support.md): DeGirum Tools API Reference Guide. Utilities to connect, run, annotate, and profile inferences.
- [Compound Models](https://docs.degirum.com/degirum-tools/compound_models.md): DeGirum Tools API Reference Guide. Chain multiple models in cropping, combining or async flows.
- [Tile Compound Models](https://docs.degirum.com/degirum-tools/tile_compound_models.md): DeGirum Tools API Reference Guide. Process large images by tiling and combining model results.
- [Streams](https://docs.degirum.com/degirum-tools/streams.md): DeGirum Tools API Reference Guide. Streaming toolkit for building multi-threaded pipelines.
- [Streams Base](https://docs.degirum.com/degirum-tools/streams/streams_base.md): DeGirum Tools API Reference Guide. Defines Stream, Gizmo and Composition core classes.
- [Streams Gizmos](https://docs.degirum.com/degirum-tools/streams/streams_gizmos.md): DeGirum Tools API Reference Guide. Reusable gizmos for video, inference, display, etc.
- [Analyzers](https://docs.degirum.com/degirum-tools/analyzers.md): DeGirum Tools API Reference Guide. Abstract base for result analyzers and overlays.
- [Zone Counter](https://docs.degirum.com/degirum-tools/analyzers/zone_count.md): DeGirum Tools API Reference Guide. Count objects present in polygonal zones.
- [Object Tracker](https://docs.degirum.com/degirum-tools/analyzers/object_tracker.md): DeGirum Tools API Reference Guide. Track objects across frames.
- [Object Selector](https://docs.degirum.com/degirum-tools/analyzers/object_selector.md): DeGirum Tools API Reference Guide. Select relevant objects while running inference.
- [Line Counter](https://docs.degirum.com/degirum-tools/analyzers/line_count.md): DeGirum Tools API Reference Guide. Count objects as they cross virtual lines.
- [Event Detector](https://docs.degirum.com/degirum-tools/analyzers/event_detector.md): DeGirum Tools API Reference Guide. Convert analyzer outputs into high-level events.
- [Notifier](https://docs.degirum.com/degirum-tools/analyzers/notifier.md): DeGirum Tools API Reference Guide. Trigger notifications when events occur.
- [Clip Saver](https://docs.degirum.com/degirum-tools/analyzers/clip_saver.md): DeGirum Tools API Reference Guide. Record video clips around trigger events.
- [Scene Cut Detector](https://docs.degirum.com/degirum-tools/analyzers/scene_cut_detector.md): DeGirum Tools API Reference Guide. Detect scene cuts in video for use with ObjectTracker and other analyzers.
- [Support Modules](https://docs.degirum.com/degirum-tools/support.md): DeGirum Tools API Reference Guide. Overview of Support modules included in DeGirum Tools.
- [Audio Support](https://docs.degirum.com/degirum-tools/support/audio_support.md): DeGirum Tools API Reference Guide. Open microphone or file-based audio streams.
- [Model Evaluation Support](https://docs.degirum.com/degirum-tools/support/eval_support.md): DeGirum Tools API Reference Guide. Framework and base class for model accuracy evaluation.
- [Math Support](https://docs.degirum.com/degirum-tools/support/math_support.md): DeGirum Tools API Reference Guide. Geometry, NMS, tiling and FIR filter helper functions.
- [Object Storage Support](https://docs.degirum.com/degirum-tools/support/object_storage_support.md): DeGirum Tools API Reference Guide. MinIO/local object-storage wrappers for file and bucket ops.
- [UI Support](https://docs.degirum.com/degirum-tools/support/ui_support.md): DeGirum Tools API Reference Guide. Lightweight display, FPS meter, timer and image-stack tools.
- [Video Support](https://docs.degirum.com/degirum-tools/support/video_support.md): DeGirum Tools API Reference Guide. Read, stream, display and save video or RTSP sources.
- [Environment Variables](https://docs.degirum.com/degirum-tools/environment-variables.md): A reference for constants and environment variables used by DeGirum Tools.
- [Remote Assets](https://docs.degirum.com/degirum-tools/remote-assets.md): Remote media assets for examples and tutorials. Thin catalog that exposes image/video samples from DeGirum's PySDK Examples as simple attributes and lists.
- [Release Notes](https://docs.degirum.com/degirum-tools/release-notes.md): This page features release notes for releases of degirum-tools. You may download degirum-tools versions listed here from PyPI.org.

## DeGirumJS

- [Overview](https://docs.degirum.com/degirumjs/readme.md): Overview of the DeGirumJS SDK and its key capabilities.
- [Getting Started](https://docs.degirum.com/degirumjs/get-started.md): Step-by-step guide for running your first inference with DeGirumJS.
- [Guides](https://docs.degirum.com/degirumjs/guides.md): Core usage guides for connection modes, model parameters, and data handling in DeGirumJS.
- [Architecture and Connection Modes](https://docs.degirum.com/degirumjs/guides/connection-modes.md): DeGirumJS offers flexible connection modes to cater to various AI inference needs, whether you're running models locally on an AI Server, entirely in the cloud, or a hybrid approach.
- [Batch Processing and Callbacks](https://docs.degirum.com/degirumjs/guides/batch-inference.md): This guide covers model.predict\_batch(), asynchronous callbacks, and how to manage the inference queue.
- [Device Management for Inference](https://docs.degirum.com/degirumjs/guides/device-management.md): Configure and switch between device types when running inference with DeGirumJS.
- [Model Parameters](https://docs.degirum.com/degirumjs/guides/model-parameters.md): Overview of model parameters available when loading or configuring models.
- [Performance and Timing Statistics](https://docs.degirum.com/degirumjs/guides/timing.md): Interpret performance and latency metrics collected during inference.
- [Preprocessing and Visual Overlays](https://docs.degirum.com/degirumjs/guides/pre-post-processing.md): Customize preprocessing and drawing parameters for DeGirumJS models.
- [Result Object Structure](https://docs.degirum.com/degirumjs/guides/result-object-structure.md): Understand the structure of prediction results returned by DeGirumJS.
- [WebCodecs Example](https://docs.degirum.com/degirumjs/guides/web-codecs-example.md): Examples for using predict\_batch with the WebCodecs API.
- [Working with Input and Output Data](https://docs.degirum.com/degirumjs/guides/input-output-data.md): Guide for the various input data formats supported by DeGirumJS for inference, as well as a detailed breakdown of the output result object structures for different model types.
- [Release Notes](https://docs.degirum.com/degirumjs/all-release-notes.md): Changelog of DeGirumJS releases.
- [API Reference Guides](https://docs.degirum.com/degirumjs/api-reference-guides.md): This page serves as an index for the API Reference Guides, providing access to detailed documentation of the classes, methods, and properties available in DeGirumJS.

## Orca

- [Overview](https://docs.degirum.com/orca/readme.md): This page provides an overview of the DeGirum Orca AI accelerator, describing its performance characteristics, support for pruned models, dedicated DRAM feature, and its flexible architecture.
- [Benchmarks](https://docs.degirum.com/orca/benchmarks.md): This page presents performance benchmark data for the DeGirum Orca AI accelerator, listing frames per second (FPS) for various models under a batch size of 1.
- [Unboxing and Installation](https://docs.degirum.com/orca/unboxing-and-installation.md): This page contains unboxing and installation instructions for the Orca M.2 Accelerator Module in M.2 form factor.
- [M.2 Setup](https://docs.degirum.com/orca/pcie.md): This page details the installation and troubleshooting procedures for the Linux kernel driver required for DeGirum Orca PCIe cards.
- [USB Setup](https://docs.degirum.com/orca/usb.md): This page describes how to install and troubleshoot DeGirum Orca devices with a USB-C interface.
- [Thermal Management](https://docs.degirum.com/orca/thermal-management.md): This topic provides guidelines for heatsink and airflow solutions to ensure the Orca M.2 module operates within safe thermal limits.
- [Tools](https://docs.degirum.com/orca/tools.md): This page describes the utilities for Orca. It covers the installation of the tools and provides usage details for utilities.

## Hailo

- [DeGirum Docs for Hailo](https://docs.degirum.com/hailo/readme.md): Start here to build and deploy edge AI on Hailo with DeGirum—quickstart, recommended journey, and cloud or local paths.
- [Before you begin](https://docs.degirum.com/hailo/basics/before-you-begin.md): Before exploring the guides and running examples, review this page to ensure you're set up for success. It covers how to install DeGirum's PySDK and DeGirum Tools with Hailo devices.
- [First inference](https://docs.degirum.com/hailo/basics/first-inference.md): Run your first inference using copy-paste-ready code.
- [Specifying a model](https://docs.degirum.com/hailo/basics/specifying-a-model.md): Before diving into code, read this page to understand how PySDK represents and handles models.
- [Discover Hailo models](https://docs.degirum.com/hailo/basics/specifying-a-model/discover-hailo-models.md): Start with precompiled Hailo models that run out of the box—learn how to pick the right variant for your device.
- [Inference setup](https://docs.degirum.com/hailo/basics/specifying-a-model/inference-setup.md): PySDK gives you flexibility in where models are stored and where inferences run. This page walks through common setups (cloud, local, and hybrid) so you can choose what fits your workflow.
- [Model properties](https://docs.degirum.com/hailo/basics/specifying-a-model/model-properties.md): See what you can tune on your model—and why it matters. This page introduces the model\_properties field and shows how to inspect, change, and group model settings by task.
- [Running inference](https://docs.degirum.com/hailo/basics/running-inference.md): Learn how to run inference with your model using simple, flexible methods—single images, video streams, or entire folders.
- [Images](https://docs.degirum.com/hailo/basics/running-inference/images.md): Run inference on a single image using a URL, file path, or NumPy array. This page shows how to use each input type with a loaded model.
- [Videos](https://docs.degirum.com/hailo/basics/running-inference/videos.md): Learn how to run real-time inference on video streams using predict\_stream. This page covers video files, webcams, and RTSP sources—all with minimal setup.
- [Folders](https://docs.degirum.com/hailo/basics/running-inference/folders.md): Run inference on entire image folders with predict\_dir, streaming results efficiently from flat or nested directories.
- [Inference results](https://docs.degirum.com/hailo/basics/inference-results.md): Understand the structure and purpose of the InferenceResults object returned by model inference. Learn how each field supports visualization, inspection, saving, or real-time streaming.
- [Inspecting results](https://docs.degirum.com/hailo/basics/inference-results/inspecting-results.md): Understand the structure of PySDK inference results so you can inspect labels, scores, and metadata before visualizing, saving, or streaming them.
- [Visualizing results](https://docs.degirum.com/hailo/basics/inference-results/visualizing-results.md): Learn how to view original frames, model-ready tensors, and overlay images returned in each inference result.
- [Saving results](https://docs.degirum.com/hailo/basics/inference-results/saving-results.md): Capture inference outputs as structured data or images so you can reuse them in downstream tools, dashboards, or datasets.
- [Streaming results](https://docs.degirum.com/hailo/basics/inference-results/streaming-results.md): Stream inference outputs in real time to displays, message buses, or remote services using PySDK result objects.
- [Measuring performance](https://docs.degirum.com/hailo/basics/measuring-performance.md): Measure latency and throughput for DeGirum models, capture per-stage timings, and apply repeatable test loops backed by consistent validation.
- [Class filtering](https://docs.degirum.com/hailo/intermediate-guides/class-filtering.md): Learn how to filter model outputs by class to reduce clutter, streamline downstream logic, and focus on what matters to your application.
- [Object tracking](https://docs.degirum.com/hailo/intermediate-guides/object-tracking.md): Learn how to track objects across video frames using degirum\_tools.ObjectTracker. This guide explains how to assign persistent IDs to detections, reduce flicker, and extract motion-based analytics.
- [Zone-based counting](https://docs.degirum.com/hailo/intermediate-guides/zone-based-counting.md): Count detections inside polygonal zones—ideal for traffic, retail, and other analytics.
- [Tiling](https://docs.degirum.com/hailo/intermediate-guides/tiling.md): Boost small-object detection using tiling. Learn four strategies to tile, detect, and merge results effectively in PySDK.
- [Model properties](https://docs.degirum.com/hailo/intermediate-guides/model-properties.md): Tune model properties to balance accuracy, latency, and visualization by adjusting preprocessing, hardware selection, and postprocessing settings.
- [Device selection](https://docs.degirum.com/hailo/intermediate-guides/model-properties/device-selection.md): Choose the device type (runtime + hardware) your model runs on. This guide covers supported types, discovery, deterministic vs. fallback selection, and pinning specific cards.
- [Input preprocessing](https://docs.degirum.com/hailo/intermediate-guides/model-properties/input-preprocessing.md): Tune how input data is resized, cropped, padded, and color-converted before reaching the model—so it matches training assumptions and avoids silent accuracy loss.
- [Output postprocessing](https://docs.degirum.com/hailo/intermediate-guides/model-properties/output-postprocessing.md): Understand how to fine-tune postprocessing parameters to control model output—filter predictions, apply suppression, and reduce clutter for more usable results.
- [Image overlay](https://docs.degirum.com/hailo/intermediate-guides/model-properties/image-overlay.md): Customize how predictions are visualized on the output image—control labels, colors, line thickness, blur, and other overlay settings without affecting model results.
- [Custom video source](https://docs.degirum.com/hailo/advanced-guides/custom-video-source.md): Plug custom video sources into PySDK using predict\_batch—ideal for cameras, SDKs, GStreamer, and advanced use cases needing per-frame control or metadata.
- [AI Server](https://docs.degirum.com/hailo/advanced-guides/ai-server.md): Learn how to use the DeGirum AI Server to efficiently host your hardware with local or cloud models.
- [Inference with cloud models](https://docs.degirum.com/hailo/advanced-guides/ai-server/ai-server-inference-with-cloud-models.md): Run inference on a local AI server while fetching models from DeGirum’s public cloud zoo—ideal for hybrid setups where compute is local, but model access is remote.
- [Inference with local models](https://docs.degirum.com/hailo/advanced-guides/ai-server/ai-server-inference-with-local-models.md): Learn how to run inference using locally stored models on a DeGirum AI Server, whether the server runs on the same machine as the client or remotely over the network.
- [PySDK examples](https://docs.degirum.com/hailo/external-resources/pysdk-examples.md): Browse runnable PySDK notebooks—ranging from first inference to multi-host and benchmarking—plus Colab launchers and end-to-end demos that complement these guides.
- [Hailo examples](https://docs.degirum.com/hailo/external-resources/hailo-examples.md): Explore the DeGirum hailo\_examples repository for runnable notebooks, environment setup, and quick validation tailored to Hailo-8 and Hailo-8L—so you can get to a known-good baseline fast.

## Axelera AI

- [DeGirum Docs for Axelera AI](https://docs.degirum.com/axelera/readme.md): Start here to build and deploy edge AI on Axelera Metis with DeGirum—quickstart, recommended journey, and cloud or local paths.
- [Before you begin](https://docs.degirum.com/axelera/basics/before-you-begin.md): Before exploring the guides and running examples, review this page to ensure you're set up for success. It covers how to install DeGirum's PySDK and DeGirum Tools with Axelera devices.
- [First inference](https://docs.degirum.com/axelera/basics/first-inference.md): Run your first inference using copy-paste-ready code.
- [Specifying a model](https://docs.degirum.com/axelera/basics/specifying-a-model.md): Before diving into code, read this page to understand how PySDK represents and handles models.
- [Discover Axelera models](https://docs.degirum.com/axelera/basics/specifying-a-model/discover-axelera-models.md): Start with precompiled Axelera models that run out of the box—learn how to pick the right variant for your device.
- [Inference setup](https://docs.degirum.com/axelera/basics/specifying-a-model/inference-setup.md): PySDK gives you flexibility in where models are stored and where inferences run. This page walks through common setups (cloud, local, and hybrid) so you can choose what fits your workflow.
- [Model properties](https://docs.degirum.com/axelera/basics/specifying-a-model/model-properties.md): See what you can tune on your model—and why it matters. This page introduces the model\_properties field and shows how to inspect, change, and group model settings by task.
- [Running inference](https://docs.degirum.com/axelera/basics/running-inference.md): Learn how to run inference with your model using simple, flexible methods—single images, video streams, or entire folders.
- [Images](https://docs.degirum.com/axelera/basics/running-inference/images.md): Run inference on a single image using a URL, file path, or NumPy array. This page shows how to use each input type with a loaded model.
- [Videos](https://docs.degirum.com/axelera/basics/running-inference/videos.md): Learn how to run real-time inference on video streams using predict\_stream. This page covers video files, webcams, and RTSP sources—all with minimal setup.
- [Folders](https://docs.degirum.com/axelera/basics/running-inference/folders.md): Run inference on entire image folders with predict\_dir, streaming results efficiently from flat or nested directories.
- [Inference results](https://docs.degirum.com/axelera/basics/inference-results.md): Understand the structure and purpose of the InferenceResults object returned by model inference. Learn how each field supports visualization, inspection, saving, or real-time streaming.
- [Inspecting results](https://docs.degirum.com/axelera/basics/inference-results/inspecting-results.md): Understand the structure of PySDK inference results so you can inspect labels, scores, and metadata before visualizing, saving, or streaming them.
- [Visualizing results](https://docs.degirum.com/axelera/basics/inference-results/visualizing-results.md): Learn how to view original frames, model-ready tensors, and overlay images returned in each inference result.
- [Saving results](https://docs.degirum.com/axelera/basics/inference-results/saving-results.md): Capture inference outputs as structured data or images so you can reuse them in downstream tools, dashboards, or datasets.
- [Streaming results](https://docs.degirum.com/axelera/basics/inference-results/streaming-results.md): Stream inference outputs in real time to displays, message buses, or remote services using PySDK result objects.
- [Measuring performance](https://docs.degirum.com/axelera/basics/measuring-performance.md): Measure latency and throughput for DeGirum models, capture per-stage timings, and apply repeatable test loops backed by consistent validation.
- [Class filtering](https://docs.degirum.com/axelera/intermediate-guides/class-filtering.md): Learn how to filter model outputs by class to reduce clutter, streamline downstream logic, and focus on what matters to your application.
- [Object tracking](https://docs.degirum.com/axelera/intermediate-guides/object-tracking.md): Learn how to track objects across video frames using degirum\_tools.ObjectTracker. This guide explains how to assign persistent IDs to detections, reduce flicker, and extract motion-based analytics.
- [Zone-based counting](https://docs.degirum.com/axelera/intermediate-guides/zone-based-counting.md): Count detections inside polygonal zones—ideal for traffic, retail, and other analytics.
- [Tiling](https://docs.degirum.com/axelera/intermediate-guides/tiling.md): Boost small-object detection using tiling. Learn four strategies to tile, detect, and merge results effectively in PySDK.
- [Model properties](https://docs.degirum.com/axelera/intermediate-guides/model-properties.md): Tune model properties to balance accuracy, latency, and visualization by adjusting preprocessing, hardware selection, and postprocessing settings.
- [Device selection](https://docs.degirum.com/axelera/intermediate-guides/model-properties/device-selection.md): Choose the device type (runtime + hardware) your model runs on. This guide covers supported types, discovery, deterministic vs. fallback selection, and pinning specific cards.
- [Input preprocessing](https://docs.degirum.com/axelera/intermediate-guides/model-properties/input-preprocessing.md): Tune how input data is resized, cropped, padded, and color-converted before reaching the model—so it matches training assumptions and avoids silent accuracy loss.
- [Output postprocessing](https://docs.degirum.com/axelera/intermediate-guides/model-properties/output-postprocessing.md): Understand how to fine-tune postprocessing parameters to control model output—filter predictions, apply suppression, and reduce clutter for more usable results.
- [Image overlay](https://docs.degirum.com/axelera/intermediate-guides/model-properties/image-overlay.md): Customize how predictions are visualized on the output image—control labels, colors, line thickness, blur, and other overlay settings without affecting model results.
- [Custom video source](https://docs.degirum.com/axelera/advanced-guides/custom-video-source.md): Plug custom video sources into PySDK using predict\_batch—ideal for cameras, SDKs, GStreamer, and advanced use cases needing per-frame control or metadata.
- [AI Server](https://docs.degirum.com/axelera/advanced-guides/ai-server.md): Learn how to use the DeGirum AI Server to efficiently host your hardware with local or cloud models.
- [Inference with cloud models](https://docs.degirum.com/axelera/advanced-guides/ai-server/ai-server-inference-with-cloud-models.md): Run inference on a local AI server while fetching models from DeGirum’s public cloud zoo—ideal for hybrid setups where compute is local, but model access is remote.
- [Inference with local models](https://docs.degirum.com/axelera/advanced-guides/ai-server/ai-server-inference-with-local-models.md): Learn how to run inference using locally stored models on a DeGirum AI Server, whether the server runs on the same machine as the client or remotely over the network.
- [PySDK examples](https://docs.degirum.com/axelera/external-resources/pysdk-examples.md): Browse runnable PySDK notebooks—ranging from first inference to multi-host and benchmarking—plus Colab launchers and end-to-end demos that complement these guides.
- [Axelera examples](https://docs.degirum.com/axelera/external-resources/axelera-examples.md): Explore the DeGirum axelera\_examples repository for runnable notebooks, environment setup, and validation tailored to Metis hardware.

## Face Recognition

- [Introduction](https://docs.degirum.com/face-recognition/index.md)
- [Installation & Setup](https://docs.degirum.com/face-recognition/getting-started/installation.md)
- [Basic Concepts](https://docs.degirum.com/face-recognition/getting-started/basic-concepts.md)
- [Face Recognizer](https://docs.degirum.com/face-recognition/guides/overview.md)
- [Configuration](https://docs.degirum.com/face-recognition/guides/overview/configuration.md)
- [Methods](https://docs.degirum.com/face-recognition/guides/overview/methods.md)
- [Face Tracker](https://docs.degirum.com/face-recognition/guides/overview-1.md)
- [Configuration](https://docs.degirum.com/face-recognition/guides/overview-1/configuration.md)
- [Methods](https://docs.degirum.com/face-recognition/guides/overview-1/methods.md)
- [Face Clip Manager](https://docs.degirum.com/face-recognition/guides/overview-2.md)
- [Configuration](https://docs.degirum.com/face-recognition/guides/overview-2/configuration.md)
- [Methods](https://docs.degirum.com/face-recognition/guides/overview-2/methods.md)
- [Database](https://docs.degirum.com/face-recognition/guides/overview-3.md)
- [Configuration](https://docs.degirum.com/face-recognition/guides/overview-3/configuration.md)
- [Methods](https://docs.degirum.com/face-recognition/guides/overview-3/methods.md)
- [Face Recognition Models](https://docs.degirum.com/face-recognition/reference/models.md)
- [Face Data Classes](https://docs.degirum.com/face-recognition/reference/face-recognition-result.md)
- [Face Filters](https://docs.degirum.com/face-recognition/reference/face-filters.md)
- [Object Storage Configuration](https://docs.degirum.com/face-recognition/reference/storage-config.md)
- [Configs](https://docs.degirum.com/face-recognition/complete-api-reference/configs.md): DeGirum Face API Reference. Configuration dataclasses and YAML loaders for recognition and tracking flows.
- [Discovery](https://docs.degirum.com/face-recognition/complete-api-reference/discovery.md): DeGirum Face API Reference. Hardware discovery helpers built on the embedded model registry.
- [Face Data](https://docs.degirum.com/face-recognition/complete-api-reference/face_data.md): DeGirum Face API Reference. Face metadata utilities plus LanceDB record structures.
- [Face Tracking](https://docs.degirum.com/face-recognition/complete-api-reference/face_tracking.md): DeGirum Face API Reference. Pipeline orchestration for enrollment, recognition, and alert handling.
- [Face Tracking Gizmos](https://docs.degirum.com/face-recognition/complete-api-reference/face_tracking_gizmos.md): DeGirum Face API Reference. Composable gizmos powering the NiceGUI tracking experience.
- [Face Utils](https://docs.degirum.com/face-recognition/complete-api-reference/face_utils.md): DeGirum Face API Reference. Shared math, filtering, and post-processing helpers.
- [Logging Config](https://docs.degirum.com/face-recognition/complete-api-reference/logging_config.md): DeGirum Face API Reference. Centralized logging setup for the face pipelines.
- [ReID Database](https://docs.degirum.com/face-recognition/complete-api-reference/reid_database.md): DeGirum Face API Reference. Re-identification database APIs backed by LanceDB.
- [Community & Support](https://docs.degirum.com/face-recognition/resources/resources.md)

## Vehicle Analytics

- [Introduction](https://docs.degirum.com/vehicle-analytics/index.md)
- [Installation & Setup](https://docs.degirum.com/vehicle-analytics/getting-started/installation.md)
- [Basic Concepts](https://docs.degirum.com/vehicle-analytics/getting-started/basic-concepts.md)
- [LP Recognizer](https://docs.degirum.com/vehicle-analytics/guides/overview.md)
- [Configuration](https://docs.degirum.com/vehicle-analytics/guides/overview/configuration.md)
- [Methods](https://docs.degirum.com/vehicle-analytics/guides/overview/methods.md)
- [LP Tracker](https://docs.degirum.com/vehicle-analytics/guides/overview-1.md)
- [Configuration](https://docs.degirum.com/vehicle-analytics/guides/overview-1/configuration.md)
- [Methods](https://docs.degirum.com/vehicle-analytics/guides/overview-1/methods.md)
- [LPR Models](https://docs.degirum.com/vehicle-analytics/reference/models.md)
- [LPR Data Classes](https://docs.degirum.com/vehicle-analytics/reference/lpr-data.md)
- [Vehicle Filters](https://docs.degirum.com/vehicle-analytics/reference/vehicle-filters.md)
- [Community & Support](https://docs.degirum.com/vehicle-analytics/resources/resources.md)

## Resources

- [External Links](https://docs.degirum.com/resources/readme.md): See below for external links to DeGirum AI Hub, GitHub, DeGirum Community, and more.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on a page URL with the `ask` query parameter:
```
GET https://docs.degirum.com/readme.md?ask=<question>
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.