The ZED SDK allows you to add depth, motion sensing and spatial AI to your application. Available as a standalone installer, it includes applications, tools and sample projects with source code.
Blackwell GPUs must use CUDA 12 or newer
ZED SDK 5.1 introduces a fully reworked image capture and recording pipeline delivering stability, maximum performance, and reliability. It adds Jetson Thor support, delivering up to 2.5× higher performance. An extra-close depth range is now supported. Streaming latency is significantly reduced, ensuring smoother real-time operation. A redesigned Virtual Stereo workflow for ZED X One now achieves up to 60% lower CPU usage. OpenCV fisheye calibration is now supported natively, and the new ZED Studio app unifies camera, stream, and SVO management. Improved Positional Tracking boosts robustness, introduces a , and refines . Expanded support for , , , and makes faster, more stable, and ready for the next generation of robotics and spatial AI.
For older releases and changelog, see the ZED SDK release archive.
Oct 23, 2025
sl::InputType::setVirtualStereoFromCameraIDs or sl::InputType::setVirtualStereoFromSerialNumbers. It shows significant performance optimization: CPU usage decreased by 60%, making this version 2.5x more CPU efficient than ZED_MediaServer, even in IPC mode. Refer to the sample virtual stereo/cpp for an example on how to use itsl::ERROR_CODE::POTENTIAL_CALIBRATION_ISSUE returned by the sl::Camera::open function, if camera calibration is poor and may require a recalibration. To verify this, open ZED_DepthViewer and perform a visual check on the scene (is the depth image full, are planes flat, are objects' shapes correct, …).getSensorsDataBatch() to retrieve all high-frequency sensor data associated with the latest grabbed frame.sl::VIEW to handle more color conventions. Added additional 3-channel (BGR) and GRAY color modes. DEPTH and CONFIDENCE views are now retrieved in color.VIDEO_SETTINGS reset calls. Now all the settings can be reset with the base function sl::Camera::setCameraSettings(setting, sl::VIDEO_SETTINGS_VALUE_AUTO), for sl::Camera and sl::CameraOne (in live and streaming mode).sl::InputType::setFromCameraID and bus_type from sl::InputType::setFromSerialNumber. Also removed the enum CAMERA_TYPE. Now, objects Camera and CameraOne handle the camera and bus types automatically.sdk_verbose of InitParametersOne to 1 to match InitParameters.sl::CameraOne. On some camera the rectified field of view was much lower than expected. To revert to the previous rectification, set the environment variable: ZED_SDK_OLD_FOV_COMPUTE=1DRIVER_FAILURE in the enum sl::ERROR_CODE. This error code can be returned when the driver's initialization has failed. When using gmsl cameras, it is then recommended to restart it with sudo systemctl restart zed_x_daemon.service.ZED_SDK_STREAM_VERSION=1. Important: The Receiver must match or exceed the sender's version for compatibility.AUTO_ANALOG_GAIN_RANGE, AUTO_DIGITAL_GAIN_RANGE, and AUTO_EXPOSURE_TIME_RANGE camera control values in streaming.PositionalTrackingParameters::enable_2d_ground_mode setting. This mode constrains positional tracking to two dimensions, recommended for grounded robots and vehicles.SPATIAL_MEMORY status in GEN3 to indicate whether tracking is active and relocated in a pre-mapped environment, performing loop closure, or relocating.getPosition(), which now returns sl::POSITIONAL_TRACKING::OK instead of previously returning sl::POSITIONAL_TRACKING::UNAVAILABLE.sl::Camera to sl::Fusion in the network configuration.sl.Mat.get_data() now supports explicit error and clearer warnings. Added instructions to set up and validate CuPy integration with PyZed.sl.FusionConfiguration in the Python wrapper.GNSS_STATUS in both GNSSData and FusedPositionalTrackingStatus.--cimud of the ZEDCalibration tool to now handle all GMSL cameras.hello_zed_gpu.py script in tutorials/tutorial1 - hello ZED/python to validate the usage of CuPy with the ZED SDK Python API.pytorch_yolov8_cupy_preproc, a YOLOv8 inference CuPy optimized sample. Also updated default weights to yolo11m.pt (previously it was yolov8m.pt)..area files, allowing for easier management of mapped environments. Also introduced the use of SPATIAL_MEMORY for better tracking of feedback.virtual stereo/cpp for an example on how to use the new VirtualStereo API from two ZED X One cameras.