# Intel® Distribution of OpenVINO™ Toolkit **URL:** https://community.intel.com/c/developer-software-tools/intel-distribution-of-openvino-toolkit/23.md [Latest](https://community.intel.com/latest.md) · [Categories](https://community.intel.com/categories.md) · [Tags](https://community.intel.com/tags.md) --- ## [OpenVINO 2026.1 Available Now!](https://community.intel.com/t/openvino-2026-1-available-now/137897)
**Author:** [@Hari\_B\_Intel](https://community.intel.com/u/Hari_B_Intel)\ **Replies:** 0\ **Last updated:** [April 13, 2026, 7:39am UTC](https://community.intel.com/t/openvino-2026-1-available-now/137897 "2026-04-13T07:39:41Z")
We are excited to introduce OpenVINO™ 2026.1! Building on the advancements of the previous release, this version brings refined performance enhancements, expanded model compatibility, and further improvements to GenAI w… --- ## [OpenVINO 2026.4 Available Now!](https://community.intel.com/t/openvino-2026-4-available-now/143775)
**Author:** [@Hari\_B\_Intel](https://community.intel.com/u/Hari_B_Intel)\ **Replies:** 0\ **Last updated:** [September 18, 2026, 4:27pm UTC](https://community.intel.com/t/openvino-2026-4-available-now/143775 "2026-09-18T16:27:27Z")
We are excited to introduce OpenVINO™ 2026.4! This release expands GenAI, multimodal AI, and speech AI capabilities with broader model coverage, advanced speculative decoding techniques, and enhanced deployment flexibil… --- ## [OpenVINO 2026.3 Available Now!](https://community.intel.com/t/openvino-2026-3-available-now/143587)
**Author:** [@Hari\_B\_Intel](https://community.intel.com/u/Hari_B_Intel)\ **Replies:** 0\ **Last updated:** [September 2, 2026, 6:53pm UTC](https://community.intel.com/t/openvino-2026-3-available-now/143587 "2026-09-02T18:53:49Z")
We are excited to introduce OpenVINO™ 2026.3! This release delivers broader GenAI and multimodal AI support, expanded model and framework compatibility, and significant performance and memory optimizations. New capabili… --- ## [About the Intel® Distribution of OpenVINO™ Toolkit category](https://community.intel.com/t/about-the-intel-distribution-of-openvino-toolkit-category/237)
**Author:** [@sam\_grogan](https://community.intel.com/u/sam_grogan)\ **Replies:** 0\ **Last updated:** [June 23, 2026, 5:32pm UTC](https://community.intel.com/t/about-the-intel-distribution-of-openvino-toolkit-category/237 "2026-06-23T17:32:25Z")
Exchange implementation strategies with deep learning engineers to quantize models, optimize inference pipelines, and deploy cross-platform computer vision and generative AI workloads. --- ## [OpenVINO 2026.2 Available Now!](https://community.intel.com/t/openvino-2026-2-available-now/142664)
**Author:** [@Hari\_B\_Intel](https://community.intel.com/u/Hari_B_Intel)\ **Replies:** 0\ **Last updated:** [June 11, 2026, 8:18am UTC](https://community.intel.com/t/openvino-2026-2-available-now/142664 "2026-06-11T08:18:42Z")
We are excited to introduce OpenVINO™ 2026.2! Building on the previous release, this version brings expanded GenAI capabilities, broader model and framework support, and enhanced performance optimizations, including fas… --- ## [OpenVino GenAI LLM benchmark not able to use multiple B60 GPU for inference](https://community.intel.com/t/openvino-genai-llm-benchmark-not-able-to-use-multiple-b60-gpu-for-inference/133673)
**Author:** [@J-N-ch](https://community.intel.com/u/J-N-ch)\ **Replies:** 3\ **Last updated:** [November 7, 2026, 10:25am UTC](https://community.intel.com/t/openvino-genai-llm-benchmark-not-able-to-use-multiple-b60-gpu-for-inference/133673 "2026-11-07T10:25:45Z")
Hi, When I try to run a benchmark of OpenVINO GenAI with multiple Intel Arc B60 GPU, I got RuntimeError as below: (B60\_LLM\_env) PS C:\\Users\\andre\\B60\_LLM\_env\\openvino.genai\\tools\\llm\_bench\> python ./benchmark.py -m mode… --- ## [Qwen3.5-4B INT8 IR fails on Intel AI Boost NPU37XX due to dynamic hybrid KV cache shapes](https://community.intel.com/t/qwen3-5-4b-int8-ir-fails-on-intel-ai-boost-npu37xx-due-to-dynamic-hybrid-kv-cache-shapes/144338)
**Author:** [@Jianmiao坚喵](https://community.intel.com/u/Jianmiao%E5%9D%9A%E5%96%B5)\ **Replies:** 2\ **Last updated:** [October 26, 2026, 6:10pm UTC](https://community.intel.com/t/qwen3-5-4b-int8-ir-fails-on-intel-ai-boost-npu37xx-due-to-dynamic-hybrid-kv-cache-shapes/144338 "2026-10-26T18:10:22Z")
OS: Windows Device: Intel(R) AI Boost / NPU37XX NPU driver: 32.0.100.5540 OpenVINO: 2026.4.1 NPU compiler version: 0x80003 Model: Qwen3.5-4B locally exported INT8 OpenVINO IR Reproduction: core = ov.Core() model … --- ## [LLM model at Visual Studio Code using OpenVINO](https://community.intel.com/t/llm-model-at-visual-studio-code-using-openvino/144161)
**Author:** [@Maxym](https://community.intel.com/u/Maxym)\ **Replies:** 1\ **Last updated:** [October 22, 2026, 4:17pm UTC](https://community.intel.com/t/llm-model-at-visual-studio-code-using-openvino/144161 "2026-10-22T16:17:12Z")
What is most correct way to run LLM model on Intel iGPU/NPU and then integrate it to Visual Studio Code to work in Agentic mode. And what is most appropiate way to run LLM for this task? Intel 155h (Intel Arc iGPU) 32g… --- ## [Community Project: Nim Bindings for the OpenVINO Runtime C API](https://community.intel.com/t/community-project-nim-bindings-for-the-openvino-runtime-c-api/144425)
**Author:** [@system](https://community.intel.com/u/system)\ **Replies:** 2\ **Last updated:** [September 30, 2026, 1:44am UTC](https://community.intel.com/t/community-project-nim-bindings-for-the-openvino-runtime-c-api/144425 "2026-09-30T01:44:34Z")
Notice: The original post is no longer available. The replies to the discussion are preserved below. --- ## [Looking for archived legacy file l\_openvino\_toolkit\_fpga\_p\_2019.3.376.tgz](https://community.intel.com/t/looking-for-archived-legacy-file-l-openvino-toolkit-fpga-p-2019-3-376-tgz/143812)
**Author:** [@dmscot](https://community.intel.com/u/dmscot)\ **Replies:** 1\ **Last updated:** [September 23, 2026, 7:11am UTC](https://community.intel.com/t/looking-for-archived-legacy-file-l-openvino-toolkit-fpga-p-2019-3-376-tgz/143812 "2026-09-23T07:11:09Z")
"I am a hobbyist/independent developer working with an older FPGA board. The old registration page is down. Could support provide a temporary link to the 2019 R3 package archive? --- ## [OpoenVINO2020.2 FPGA support](https://community.intel.com/t/opoenvino2020-2-fpga-support/89120)
**Author:** [@Kobu\_Intel](https://community.intel.com/u/Kobu_Intel)\ **Replies:** 4\ **Last updated:** [September 22, 2026, 7:28am UTC](https://community.intel.com/t/opoenvino2020-2-fpga-support/89120 "2026-09-22T07:28:39Z")
Hi, OpenVINO release statement may need to update, FPGA Bit Stream is in 2020.2 package, it’s assumed no reason to not recommend 20202. for FPGA. thanks, BR/Kobu https://software.intel.com/en-us/articles/OpenVINO-RelNo… --- ## [Object detection engine on Intel CPU via OpenVINO: 130-150 FPS and 3 MB heap on i5-11400](https://community.intel.com/t/object-detection-engine-on-intel-cpu-via-openvino-130-150-fps-and-3-mb-heap-on-i5-11400/143673)
**Author:** [@olesha-ai](https://community.intel.com/u/olesha-ai)\ **Replies:** 5\ **Last updated:** [September 15, 2026, 12:44pm UTC](https://community.intel.com/t/object-detection-engine-on-intel-cpu-via-openvino-130-150-fps-and-3-mb-heap-on-i5-11400/143673 "2026-09-15T12:44:44Z")
Hi everyone,I want to share my practical case. Below is the full technical breakdown:Practical Case: 130-150 FPS and 3 MB of Memory on a Core i5 for YOLO Inference Without a GPU via OpenVINOThis article was inspired by m… --- ## [Requesting urgent approval of Instance for AI Infra Summit Hackathon](https://community.intel.com/t/requesting-urgent-approval-of-instance-for-ai-infra-summit-hackathon/143686)
**Author:** [@Ashutosh-Pandey](https://community.intel.com/u/Ashutosh-Pandey)\ **Replies:** 1\ **Last updated:** [September 14, 2026, 4:55pm UTC](https://community.intel.com/t/requesting-urgent-approval-of-instance-for-ai-infra-summit-hackathon/143686 "2026-09-14T16:55:28Z")
My team is an online participant in the AI Infra Summit Hackathon which is supported by Intel, and the problem statement requires benchmarking our solution on an Intel Ultra Series 2 or Intel Ultra Series 3 machine. We o… --- ## [Intel Linux AI Distro . Hou can i submit our O.S](https://community.intel.com/t/intel-linux-ai-distro-hou-can-i-submit-our-o-s/143062)
**Author:** [@FabioHenrique](https://community.intel.com/u/FabioHenrique)\ **Replies:** 2\ **Last updated:** [August 17, 2026, 3:14pm UTC](https://community.intel.com/t/intel-linux-ai-distro-hou-can-i-submit-our-o-s/143062 "2026-08-17T15:14:53Z")
Hello We have developed a fully functional operating system optimized for AI, edge computing, and Intel-based platforms. Throughout the development and testing process, we have focused on performance, responsiveness, ha… --- ## [Running official OpenVINO model with target\_device NPU leads to error, but GPU works](https://community.intel.com/t/running-official-openvino-model-with-target-device-npu-leads-to-error-but-gpu-works/142843)
**Author:** [@nicolazilio](https://community.intel.com/u/nicolazilio)\ **Replies:** 4\ **Last updated:** [August 17, 2026, 12:32pm UTC](https://community.intel.com/t/running-official-openvino-model-with-target-device-npu-leads-to-error-but-gpu-works/142843 "2026-08-17T12:32:43Z")
Dear all, I have a problem running official OpenVINO models on an Intel NPU on Windows. All my Intel Driver are up-to-date and I am running the latest version of OpenVINO Model Server Intel NPU: 32.0.100.4778\In… --- ## [Product: Intel NUC8 (NUC8v5PNH, NUC8v5PNK, NUC8i3BEH, NUC8i3BEK) OS: AlmaLinux / RHEL 9.8 (kernel 5.](https://community.intel.com/t/product-intel-nuc8-nuc8v5pnh-nuc8v5pnk-nuc8i3beh-nuc8i3bek-os-almalinux-rhel-9-8-kernel-5/143197)
**Author:** [@sri467-cyara](https://community.intel.com/u/sri467-cyara)\ **Replies:** 2\ **Last updated:** [August 3, 2026, 3:04pm UTC](https://community.intel.com/t/product-intel-nuc8-nuc8v5pnh-nuc8v5pnk-nuc8i3beh-nuc8i3bek-os-almalinux-rhel-9-8-kernel-5/143197 "2026-08-03T15:04:07Z")
Issue: NUC8 hardware fails to boot AlmaLinux/RHEL 9.8 kernel due to TPM/Intel PTT initialisation failure: \[Firmware Bug\]: TPM interrupt not working, polling instead VFS: Unable to mount root fs on unknown-block(0,0) … --- ## [Building from source openvino on Ubuntu 24.4](https://community.intel.com/t/building-from-source-openvino-on-ubuntu-24-4/137983)
**Author:** [@DH55T5](https://community.intel.com/u/DH55T5)\ **Replies:** 10\ **Last updated:** [August 3, 2026, 4:40am UTC](https://community.intel.com/t/building-from-source-openvino-on-ubuntu-24-4/137983 "2026-08-03T04:40:16Z")
I’ve been trying hard to have a fully installed Openvino Toolkit from source but everytime, something is missing. Im on Ubuntu 24.4, with a Intel CPU i7 and a old IGPU so I tought it will be the best. Unfortunatly, afte… --- ## [Develop small box type 3D conversion for Medical Device by using Core Ultra or Intel i9 / Open VINO](https://community.intel.com/t/develop-small-box-type-3d-conversion-for-medical-device-by-using-core-ultra-or-intel-i9-open-vino/138310)
**Author:** [@nc9](https://community.intel.com/u/nc9)\ **Replies:** 4\ **Last updated:** [July 15, 2026, 4:41am UTC](https://community.intel.com/t/develop-small-box-type-3d-conversion-for-medical-device-by-using-core-ultra-or-intel-i9-open-vino/138310 "2026-07-15T04:41:43Z")
We try to use the below: Intel Core Ultra or Intel i9 / Open VINO Real-time inference Edge AI Medical imaging Any ideas for electrical board design and development? What do you think, Any advices? --- ## [403 Forbidden Error when accessing OpenVINO 2023 APT Repository (apt.repos.intel.com)](https://community.intel.com/t/403-forbidden-error-when-accessing-openvino-2023-apt-repository-apt-repos-intel-com/142648)
**Author:** [@LauCB](https://community.intel.com/u/LauCB)\ **Replies:** 4\ **Last updated:** [June 18, 2026, 5:33am UTC](https://community.intel.com/t/403-forbidden-error-when-accessing-openvino-2023-apt-repository-apt-repos-intel-com/142648 "2026-06-18T05:33:38Z")
Hello, Since today, June 9th, I have been unable to access the OpenVINO 2023 APT repository. The apt-get update command is consistently failing with a 403 Forbidden error. My repository configuration in Dockerfile, whi… --- ## [Unable to use HETERO Plugin with manual affinity](https://community.intel.com/t/unable-to-use-hetero-plugin-with-manual-affinity/138183)
**Author:** [@GlowingScrewdriver](https://community.intel.com/u/GlowingScrewdriver)\ **Replies:** 4\ **Last updated:** [May 21, 2026, 5:50am UTC](https://community.intel.com/t/unable-to-use-hetero-plugin-with-manual-affinity/138183 "2026-05-21T05:50:09Z")
Hi, I’m trying to run a model on OpenVINO HETERO plugin, while manually assigning node affinities to force some layers to execute on the CPU. The model in question is a transformer model from HuggingFace (Qwen3-8B)… --- ## [Nuc](https://community.intel.com/t/nuc/138245)
**Author:** [@AndrisAR20](https://community.intel.com/u/AndrisAR20)\ **Replies:** 2\ **Last updated:** [May 21, 2026, 5:47am UTC](https://community.intel.com/t/nuc/138245 "2026-05-21T05:47:27Z")
“Hello Peh, thank you for your previous help. I have an additional question — could you point me to the right Intel contact for NUC hardware recommendations and potential volume orders for an ISV accessibility device?” --- ## [Integrate](https://community.intel.com/t/integrate/138086)
**Author:** [@AndrisAR20](https://community.intel.com/u/AndrisAR20)\ **Replies:** 2\ **Last updated:** [May 8, 2026, 5:00am UTC](https://community.intel.com/t/integrate/138086 "2026-05-08T05:00:40Z")
“Hello, I am building a real-time speech recognition device on Intel hardware. How can I integrate OpenVINO to speed up Whisper ASR inference? Where should I start?” --- ## [OpenVino](https://community.intel.com/t/openvino/138095)
**Author:** [@AndrisAR20](https://community.intel.com/u/AndrisAR20)\ **Replies:** 1\ **Last updated:** [April 27, 2026, 7:52pm UTC](https://community.intel.com/t/openvino/138095 "2026-04-27T19:52:52Z")
“Hi, I’m building a real-time speech recognition device on Intel hardware. How can I integrate OpenVINO to accelerate Whisper ASR inference? Where do I start?” --- ## [Cropping a video in a dlstream pipeline](https://community.intel.com/t/cropping-a-video-in-a-dlstream-pipeline/137707)
**Author:** [@Mannberg](https://community.intel.com/u/Mannberg)\ **Replies:** 5\ **Last updated:** [April 24, 2026, 12:58am UTC](https://community.intel.com/t/cropping-a-video-in-a-dlstream-pipeline/137707 "2026-04-24T00:58:39Z")
Im using gstreamer and dlstreamer in a pipeline and I need to crop out a part of the video, for example this pipeline works: gst-launch-1.0 \\ filesrc location="$VIDEO\_FILE" ! \\ decodebin3 ! \\ videoconvert ! \\ … --- ## [Achieving Sub-10s/it Stable Diffusion Inference on Intel Core i3-1005G1 (Ice Lake) \[Benchmark\]](https://community.intel.com/t/achieving-sub-10s-it-stable-diffusion-inference-on-intel-core-i3-1005g1-ice-lake-benchmark/138038)
**Author:** [@MiniI3](https://community.intel.com/u/MiniI3)\ **Replies:** 1\ **Last updated:** [April 23, 2026, 1:48am UTC](https://community.intel.com/t/achieving-sub-10s-it-stable-diffusion-inference-on-intel-core-i3-1005g1-ice-lake-benchmark/138038 "2026-04-23T01:48:52Z")
Subject: \[Benchmark\] Achieving Sub-10s/it Stable Diffusion Inference on Intel Core i3-1005G1 (Ice Lake) Overview Following my previous report, I have further optimized my Stable Diffusion 1.5 pipeline on the Intel® Core™… --- ## [OpenVINO Model Server](https://community.intel.com/t/openvino-model-server/134057)
**Author:** [@ashish017](https://community.intel.com/u/ashish017)\ **Replies:** 3\ **Last updated:** [April 16, 2026, 7:28am UTC](https://community.intel.com/t/openvino-model-server/134057 "2026-04-16T07:28:00Z")
I am running OVMS model using chat/completions endpoint. The input is list of dictionaries: \[ {“role”: “system” : “You are an helpful assistant.”}, {“role”: “user” : “Hi! How are you?”}, {“role”: “assistant” : “.”} \] T… --- ## [OpenVINO CPU Scaling Bottleneck](https://community.intel.com/t/openvino-cpu-scaling-bottleneck/137778)
**Author:** [@AChan74](https://community.intel.com/u/AChan74)\ **Replies:** 3\ **Last updated:** [April 8, 2026, 3:12am UTC](https://community.intel.com/t/openvino-cpu-scaling-bottleneck/137778 "2026-04-08T03:12:10Z")
Hey all, I’ve opened a CPU scaling performance issue with OpenVINO and the details are on GitHub: Has anyone encountered similar issues before and solved it? Please advice. --- ## [operation compatiblity](https://community.intel.com/t/operation-compatiblity/137708)
**Author:** [@ZaidSahawneh](https://community.intel.com/u/ZaidSahawneh)\ **Replies:** 2\ **Last updated:** [April 7, 2026, 5:37am UTC](https://community.intel.com/t/operation-compatiblity/137708 "2026-04-07T05:37:49Z")
Hello, I found a browser tool that allowed me to check the support for different model operations: but it suggests to reference the documentation for something more up to date. I’m having trouble finding a clear … --- ## [Testing LLM and VLM Serving Accuracy](https://community.intel.com/t/testing-llm-and-vlm-serving-accuracy/137297)
**Author:** [@IPatriort](https://community.intel.com/u/IPatriort)\ **Replies:** 13\ **Last updated:** [March 24, 2026, 4:34am UTC](https://community.intel.com/t/testing-llm-and-vlm-serving-accuracy/137297 "2026-03-24T04:34:35Z")
Hi, i would like to ask, as i notice there is a code sample where the following will run on cpu without issue, but when i tried to swap the target device to GPU, i seem to receive request api error, how do i fix the foll… --- ## [The operator 'DetectionOutput' for iGPU & NPU failed to be compiled.](https://community.intel.com/t/the-operator-detectionoutput-for-igpu-npu-failed-to-be-compiled/136936)
**Author:** [@Byung-Hun\_Park](https://community.intel.com/u/Byung-Hun_Park)\ **Replies:** 9\ **Last updated:** [March 13, 2026, 3:15pm UTC](https://community.intel.com/t/the-operator-detectionoutput-for-igpu-npu-failed-to-be-compiled/136936 "2026-03-13T15:15:14Z")
Hi, everyone. I constructed a DNN model for object detection and am using OpenVINO 2025.4 for the inference. One strange problem occurs inferring with OpenVINO. Although there is no problem when the model is compiled a… [Next page](https://community.intel.com/c/developer-software-tools/intel-distribution-of-openvino-toolkit/23.md?page=1)