Intel® Distribution of OpenVINO™ Toolkit

753640
8/26/2026

Introduction

This package contains the Intel® Distribution of OpenVINO™ Toolkit software version 2026.3 for Linux*, Windows*, and macOS*.

Available Downloads

  • Microsoft Windows*
  • Size: 198 MB
  • SHA256: 1F94CD7DD2F3B54FE8F0D3F7F77FE0C7D5AC317AAA65AB352D6CBB0459A978B1
  • Microsoft Windows*
  • Size: 1.4 GB
  • SHA256: BDE100220774D9F1F4D40E1854779C1EE268528A53E4626350DD0C583E6D346D
  • iOS*
  • Size: 40.8 MB
  • SHA256: 3FBAD7CF6273E5ED1D23D84C9B8CBEEA770461D096B2E72D41D868D433FF8E52
  • Android*
  • Size: 40.3 MB
  • SHA256: 757EB8140039C9060A9E2668DC3885AB0DAF24FABE7ADF71A8E9E81DA148049C
  • Red Hat Linux Family*
  • Size: 50.8 MB
  • SHA256: F4BD4CCC245563E65ED892EE7DCB0A8764081BD57AAB3EC7DB6157108487B8A7
  • Ubuntu 22.04 LTS*
  • Size: 39.8 MB
  • SHA256: C5C9B4E8B44C51707A6E92F28D3B8952168BFD243403CE3919698CE804126E22
  • CentOS Linux Family*
  • Size: 71.6 MB
  • SHA256: E6FA4FE8A96046508B8F5862FA704AD6CDAC0DC4D4E67D139BF387116EB5C5B0
  • Red Hat Enterprise Linux 8.3*
  • Size: 74.4 MB
  • SHA256: EDB967509318F2EB498EFBA7F40C1F663C22E3B82D1EA57182A14C4DC2843AB0
  • Ubuntu 24.04 LTS*
  • Size: 62 MB
  • SHA256: 5865938CDB8E9854830131E3998FBBD7CDF86810196D98A041ED5C92750B9148
  • Ubuntu 22.04 LTS*
  • Size: 103.3 MB
  • SHA256: 36576CEE74F84E3986A659305FCD81FFEDB629E56D4D4ACFF875F9008DAED9F4
  • Ubuntu 24.04 LTS*
  • Size: 105.8 MB
  • SHA256: CB84D1CCDECB8BD90337EDF63E8D081A17CDE2FC72F5E9FEB213B5B2F96EB21B

Detailed Description

What’s New

  • More Gen AI coverage and frameworks integrations to minimize code changes

    • New models supported:

      • On CPU, GPU, and NPU: SmolLM3-3B, LFM2-1.2B, LFM2.5-1.2B

      • On CPU and GPU: Harrier OSS-v1-0.6B, Qwen3-8B with EAGLE-3, MiniCPM5-1B, FLUX.2-klein-4B

      • Extended to GPU and NPU: YOLO26

      • Additional models available as early releases on CPU and GPU: Qwen3-ASR, Qwen3-Omni, Gemma-3n, Qwen3-VL-Embedding-8B, Kokoro-82M

    • Support for Hugging Face Transformers v5.5, ensuring compatibility with the latest model architectures on Hugging Face

    Broader LLM model support and more model compression techniques

    • OpenVINO™ GenAI extends the EAGLE-3 speculative decoding pipeline to LLMs and VLMs, enhancing existing continuous batching and adding Top-K sampling to deliver additional token-generation speedups on CPUs, GPUs, and NPUs.

    • Lazy weight loading is enabled for IR and ONNX models to automatically select an optimal loading and compilation path, minimizing peak memory usage during model initialization.

    • Neural Network Compression Framework (NNCF) now supports FP8 quantization for ONNX models, helping developers realize FP8 performance, accuracy, and memory gains while expanding low-precision inference options for production deployments.

    • OpenVINO™ GenAI now supports three additional pipelines: Omni for multimodal workloads, ASR for speech recognition, and Embedding for multimodal embedding generation.

    More portability and performance to run AI at the edge, in the cloud or locally

    • Introducing support for Intel® Xeon® 6+ processors (formerly codenamed Clearwater Forest)

    • MoE offloading to disk enabled, allowing 30B MoE models like Qwen3-30B-A3B to run even on devices with 16 GB of memory while maintaining acceptable tokens-per-second (TPS) generation rates.

    • OpenVINO™ Model Server simplifies model deployment and unifies REST API endpoints, reducing command complexity while providing standard v1/chat/completions support for easier integration with other serving frameworks.

    • OpenVINO™ Model Server further improves stability, performance, and accuracy for LLMs such as Qwen3.5/3.6 with linear attention and extends tool-parser support to MiniCPM5-1B and LFM2.5, improving accuracy and reliability.

Get all the details. See 2026.3 release notes.

Installation instructions

You can choose how to install OpenVINO™ Runtime from Archive* according to your operating system:

What's included in the download package (Archive File)

  • Offers both C/C++ and Python APIs
  • Additionally includes code samples

 

Helpful Links

NOTE: Links open in a new window.

Disclaimers1

Product and Performance Information

1

Intel is in the process of removing non-inclusive language from our current documentation, user interfaces, and code. Please note that retroactive changes are not always possible, and some non-inclusive language may remain in older documentation, user interfaces, and code.