Link to this sectionIntel OpenVINO Export#

In this guide, we cover exporting YOLO26 models to the OpenVINO format, which can provide up to 3x CPU speedup, as well as accelerating YOLO inference on Intel GPU and NPU hardware.

OpenVINO, short for Open Visual Inference & Neural Network Optimization toolkit, is a comprehensive toolkit for optimizing and deploying AI inference models. Even though the name contains Visual, OpenVINO also supports various additional tasks including language, audio, time series, etc.

Link to this sectionUsage Examples#

The OpenVINO format supports the Export, Predict, and Validate modes. Export your model, then load the exported model to run inference or validate its accuracy on Intel CPU, integrated/discrete GPU, or NPU.

from ultralytics import YOLO

# Load a YOLO26 model
model = YOLO("yolo26n.pt")

# Export the model to OpenVINO format
model.export(format="openvino")  # creates 'yolo26n_openvino_model/'

from ultralytics import YOLO

# Load the exported OpenVINO model
model = YOLO("yolo26n_openvino_model/")

# Run inference
results = model("https://ultralytics.com/images/bus.jpg")

# Run inference on a specific device, available devices: ["intel:gpu", "intel:npu", "intel:cpu"]
results = model("https://ultralytics.com/images/bus.jpg", device="intel:gpu")

from ultralytics import YOLO

# Load the exported OpenVINO model
model = YOLO("yolo26n_openvino_model/")

# Validate accuracy on the COCO8 dataset
metrics = model.val(data="coco8.yaml")

Link to this sectionExport Arguments#

For more details about the export process, visit the Ultralytics documentation page on exporting.

Link to this sectionBenefits of OpenVINO#

1. Performance: OpenVINO delivers high-performance inference by utilizing the power of Intel CPUs, integrated and discrete GPUs, and FPGAs.
2. Support for Heterogeneous Execution: OpenVINO provides an API to write once and deploy on any supported Intel hardware (CPU, GPU, FPGA, VPU, etc.).
3. Model Optimizer: OpenVINO provides a Model Optimizer that imports, converts, and optimizes models from popular deep learning frameworks such as PyTorch, TensorFlow, TensorFlow Lite, Keras, ONNX, PaddlePaddle, and Caffe.
4. Ease of Use: The toolkit comes with more than 80 tutorial notebooks (including YOLO26 optimization) teaching different aspects of the toolkit.

Link to this sectionOpenVINO Export Structure#

When you export a model to OpenVINO format, it results in a directory containing the following:

1. XML file: Describes the network topology.
2. BIN file: Contains the weights and biases binary data.
3. Mapping file: Holds mapping of original model output tensors to OpenVINO tensor names.

You can use these files to run inference with the OpenVINO Inference Engine.

Link to this sectionUsing OpenVINO Export in Deployment#

Once your model is successfully exported to the OpenVINO format, you have two primary options for running inference:

1. Use the ultralytics package, which provides a high-level API and wraps the OpenVINO Runtime.

2. Use the native openvino package for more advanced or customized control over inference behavior.

Link to this sectionInference with Ultralytics#

The ultralytics package allows you to easily run inference using the exported OpenVINO model via the predict method. You can also specify the target device (e.g., intel:gpu, intel:npu, intel:cpu) using the device argument.

from ultralytics import YOLO

# Load the exported OpenVINO model
ov_model = YOLO("yolo26n_openvino_model/")  # the path of your exported OpenVINO model
# Run inference with the exported model
ov_model.predict(device="intel:gpu")  # specify the device you want to run inference on

This approach is ideal for fast prototyping or deployment when you don't need full control over the inference pipeline.

Link to this sectionInference with OpenVINO Runtime#

The OpenVINO Runtime provides a unified API for inference across all supported Intel hardware. It also provides advanced capabilities like load balancing across Intel hardware and asynchronous execution. For more information on running inference, refer to the YOLO26 notebooks.

Remember, you'll need the XML and BIN files as well as any application-specific settings like input size, scale factor for normalization, etc., to correctly set up and use the model with the Runtime.

In your deployment application, you would typically do the following steps:

1. Initialize OpenVINO by creating core = Core().
2. Load the model using the core.read_model() method.
3. Compile the model using the core.compile_model() function.
4. Prepare the input (image, text, audio, etc.).
5. Run inference using compiled_model(input_data).

For more detailed steps and code snippets, refer to the OpenVINO documentation or API tutorial.

Link to this sectionOpenVINO YOLO26 Benchmarks#

The Ultralytics team benchmarked YOLO26 across various model formats and precision, evaluating speed and accuracy on different Intel devices compatible with OpenVINO.

Link to this sectionIntel® Core™ Ultra#

The Intel® Core™ Ultra™ series represents a new benchmark in high-performance computing, engineered to meet the evolving demands of modern users—from gamers and creators to professionals leveraging AI. This next-generation lineup is more than a traditional CPU series; it combines powerful CPU cores, integrated high-performance GPU capabilities, and a dedicated Neural Processing Unit (NPU) within a single chip, offering a unified solution for diverse and intensive computing workloads.

At the heart of the Intel® Core Ultra™ architecture is a hybrid design that enables exceptional performance across traditional processing tasks, GPU-accelerated workloads, and AI-driven operations. The inclusion of the NPU enhances on-device AI inference, enabling faster, more efficient machine learning and data processing across a wide range of applications.

The Core Ultra™ family includes various models tailored for different performance needs, with options ranging from energy-efficient designs to high-power variants marked by the "H" designation—ideal for laptops and compact form factors that demand serious computing power. Across the lineup, users benefit from the synergy of CPU, GPU, and NPU integration, delivering remarkable efficiency, responsiveness, and multitasking capabilities.

As part of Intel's ongoing innovation, the Core Ultra™ series sets a new standard for future-ready computing. With multiple models available and more on the horizon, this series underscores Intel's commitment to delivering cutting-edge solutions for the next generation of intelligent, AI-enhanced devices.

Benchmarks below run on Intel® Core™ Ultra™ X7 358H, Intel® Core™ Ultra™ 7 258V and Intel® Core™ Ultra™ 7 155H at FP32, FP16 and INT8 precision.

Link to this sectionIntel® Core™ Ultra™ X7 358H#

Benchmarks

Integrated Intel® Arc™ GPUIntel® Panther Lake CPUIntegrated Intel® AI Boost NPU

Detailed Benchmark Results

Model	Format	Precision	Status	Size (MB)	metrics/mAP50-95(B)	Inference time (ms/im)
YOLO26n	PyTorch (CPU)	FP32	✅	5.3	0.4765	25.18
YOLO26n	OpenVINO	FP32	✅	9.6	0.4763	2.67
YOLO26n	OpenVINO	FP16	✅	5.1	0.4763	2.64
YOLO26n	OpenVINO	INT8	✅	3.2	0.4625	2.73
YOLO26s	PyTorch (CPU)	FP32	✅	19.5	0.5703	50.09
YOLO26s	OpenVINO	FP32	✅	36.7	0.5615	3.57
YOLO26s	OpenVINO	FP16	✅	18.6	0.5615	3.55
YOLO26s	OpenVINO	INT8	✅	10.0	0.547	3.09
YOLO26m	PyTorch (CPU)	FP32	✅	42.2	0.6196	135.1
YOLO26m	OpenVINO	FP32	✅	78.3	0.6168	5.64
YOLO26m	OpenVINO	FP16	✅	39.5	0.6168	5.85
YOLO26m	OpenVINO	INT8	✅	20.5	0.5994	4.14
YOLO26l	PyTorch (CPU)	FP32	✅	50.7	0.6215	169.75
YOLO26l	OpenVINO	FP32	✅	95.3	0.6206	8.14
YOLO26l	OpenVINO	FP16	✅	48.1	0.6206	8.18
YOLO26l	OpenVINO	INT8	✅	25.2	0.5999	4.67
YOLO26x	PyTorch (CPU)	FP32	✅	113.2	0.6512	407.56
YOLO26x	OpenVINO	FP32	✅	213.2	0.6569	13.11
YOLO26x	OpenVINO	FP16	✅	107.1	0.6569	13.15
YOLO26x	OpenVINO	INT8	✅	54.8	0.6374	9.24

Link to this sectionIntel® Core™ Ultra™ 7 258V#

Benchmarks

Integrated Intel® Arc™ GPUIntel® Lunar Lake CPUIntegrated Intel® AI Boost NPU

Detailed Benchmark Results

Model	Format	Precision	Status	Size (MB)	metrics/mAP50-95(B)	Inference time (ms/im)
YOLO26n	PyTorch (CPU)	FP32	✅	5.3	0.4765	31.43
YOLO26n	OpenVINO	FP32	✅	9.6	0.4762	3.57
YOLO26n	OpenVINO	FP16	✅	5.1	0.4762	3.53
YOLO26n	OpenVINO	INT8	✅	3.2	0.4625	3.65
YOLO26s	PyTorch (CPU)	FP32	✅	19.5	0.5703	60.4
YOLO26s	OpenVINO	FP32	✅	36.7	0.5616	5.02
YOLO26s	OpenVINO	FP16	✅	18.6	0.5616	5.01
YOLO26s	OpenVINO	INT8	✅	10.0	0.547	4.31
YOLO26m	PyTorch (CPU)	FP32	✅	42.2	0.6196	173.31
YOLO26m	OpenVINO	FP32	✅	78.3	0.6191	9.48
YOLO26m	OpenVINO	FP16	✅	39.5	0.6168	9.6
YOLO26m	OpenVINO	INT8	✅	20.5	0.5994	6.03
YOLO26l	PyTorch (CPU)	FP32	✅	50.7	0.6173	224.52
YOLO26l	OpenVINO	FP32	✅	95.3	0.3725	11.88
YOLO26l	OpenVINO	FP16	✅	48.1	0.6201	12.0
YOLO26l	OpenVINO	INT8	✅	25.2	0.5999	8.47
YOLO26x	PyTorch (CPU)	FP32	✅	113.2	0.6512	595.72
YOLO26x	OpenVINO	FP32	✅	213.2	0.6567	20.26
YOLO26x	OpenVINO	FP16	✅	107.1	0.6454	20.25
YOLO26x	OpenVINO	INT8	✅	54.8	0.6374	14.77

Link to this sectionIntel® Core™ Ultra™ 7 155H#

Benchmarks

Integrated Intel® Arc™ GPUIntel® Meteor Lake CPU

Detailed Benchmark Results

Model	Format	Precision	Status	Size (MB)	metrics/mAP50-95(B)	Inference time (ms/im)
YOLO26n	PyTorch (CPU)	FP32	✅	5.3	0.4765	38.77
YOLO26n	OpenVINO	FP32	✅	9.6	0.4774	9.87
YOLO26n	OpenVINO	FP16	✅	5.1	0.4774	9.84
YOLO26n	OpenVINO	INT8	✅	3.2	0.4705	5.86
YOLO26s	PyTorch (CPU)	FP32	✅	19.5	0.5703	69.54
YOLO26s	OpenVINO	FP32	✅	36.7	0.5616	17.29
YOLO26s	OpenVINO	FP16	✅	18.6	0.5616	17.06
YOLO26s	OpenVINO	INT8	✅	10.0	0.5452	10.33
YOLO26m	PyTorch (CPU)	FP32	✅	42.2	0.6196	192.22
YOLO26m	OpenVINO	FP32	✅	78.3	0.6187	34.64
YOLO26m	OpenVINO	FP16	✅	39.5	0.6187	34.75
YOLO26m	OpenVINO	INT8	✅	20.5	0.6073	15.99
YOLO26l	PyTorch (CPU)	FP32	✅	50.7	0.6215	245.62
YOLO26l	OpenVINO	FP32	✅	95.3	0.6202	43.7
YOLO26l	OpenVINO	FP16	✅	48.1	0.6202	44.65
YOLO26l	OpenVINO	INT8	✅	25.2	0.6048	20.31
YOLO26x	PyTorch (CPU)	FP32	✅	113.2	0.6512	513.06
YOLO26x	OpenVINO	FP32	✅	213.2	0.6544	80.19
YOLO26x	OpenVINO	FP16	✅	107.1	0.6544	79.83
YOLO26x	OpenVINO	INT8	✅	54.8	0.6393	35.16

Link to this sectionReproduce Our Results#

To reproduce the Ultralytics benchmarks above on all export formats run this code:

from ultralytics import YOLO

# Load a YOLO26n PyTorch model
model = YOLO("yolo26n.pt")

# Benchmark YOLO26n speed and accuracy on the COCO128 dataset for all export formats
results = model.benchmark(data="coco128.yaml")

Link to this sectionConclusion#

The benchmarking results clearly demonstrate the benefits of exporting the YOLO26 model to the OpenVINO format. Across different models and hardware platforms, the OpenVINO format consistently outperforms other formats in terms of inference speed while maintaining comparable accuracy.

The benchmarks underline the effectiveness of OpenVINO as a tool for deploying deep learning models. By converting models to the OpenVINO format, developers can achieve significant performance improvements, making it easier to deploy these models in real-world applications.

For more detailed information and instructions on using OpenVINO, refer to the official OpenVINO documentation.

Link to this sectionFAQ#

Link to this sectionHow do I export YOLO26 models to OpenVINO format?#

Exporting YOLO26 models to the OpenVINO format can significantly enhance CPU speed and enable GPU and NPU accelerations on Intel hardware. To export, you can use either Python or CLI as shown below:

from ultralytics import YOLO

# Load a YOLO26n PyTorch model
model = YOLO("yolo26n.pt")

# Export the model
model.export(format="openvino")  # creates 'yolo26n_openvino_model/'

For more information, refer to the export formats documentation.

Link to this sectionWhat are the benefits of using OpenVINO with YOLO26 models?#

Using Intel's OpenVINO toolkit with YOLO26 models offers several benefits:

1. Performance: Achieve up to 3x speedup on CPU inference and leverage Intel GPUs and NPUs for acceleration.
2. Model Optimizer: Convert, optimize, and execute models from popular frameworks like PyTorch, TensorFlow, and ONNX.
3. Ease of Use: Over 80 tutorial notebooks are available to help users get started, including ones for YOLO26.
4. Heterogeneous Execution: Deploy models on various Intel hardware with a unified API.

For detailed performance comparisons, visit our benchmarks section.

Link to this sectionHow can I run inference using a YOLO26 model exported to OpenVINO?#

After exporting a YOLO26n model to OpenVINO format, you can run inference using Python or CLI:

from ultralytics import YOLO

# Load the exported OpenVINO model
ov_model = YOLO("yolo26n_openvino_model/")

# Run inference
results = ov_model("https://ultralytics.com/images/bus.jpg")

Refer to our predict mode documentation for more details.

Link to this sectionWhy should I choose Ultralytics YOLO26 over other models for OpenVINO export?#

Ultralytics YOLO26 is optimized for real-time object detection with high accuracy and speed. Specifically, when combined with OpenVINO, YOLO26 provides:

• Up to 3x speedup on Intel CPUs
• Seamless deployment on Intel GPUs and NPUs
• Consistent and comparable accuracy across various export formats

For in-depth performance analysis, check our detailed YOLO26 benchmarks on different hardware.

Link to this sectionCan I benchmark YOLO26 models on different formats such as PyTorch, ONNX, and OpenVINO?#

Yes, you can benchmark YOLO26 models in various formats including PyTorch, TorchScript, ONNX, and OpenVINO. Use the following code snippet to run benchmarks on your chosen dataset:

from ultralytics import YOLO

# Load a YOLO26n PyTorch model
model = YOLO("yolo26n.pt")

# Benchmark YOLO26n speed and [accuracy](https://www.ultralytics.com/glossary/accuracy) on the COCO8 dataset for all export formats
results = model.benchmark(data="coco8.yaml")

For detailed benchmark results, refer to our benchmarks section and export formats documentation.