Thanks to the state-of-the-art Intel AI stack, both hardware platforms and software tools including Intel extension for PyTorch, great performance improvements (1.61x in latency and 2.91x in throughput) were achieved for our Vision-based Motion Capture System.
Tailor Zhang Project Manager – West China PI Tech
Not only did we get a significant boost in the inference speed using OpenVINO, we now have the capability to leverage Intel Xeon Scalable CPUs to do inferences in real-time.
Ankit Narayan Singh Co-Founder & CTO, ParallelDots
Leveraging the Intel optimized distribution for Python, OpenVINO toolkit on 3rd Gen Intel Xeon Scalable Processors, we are able to speed up our SubtleMR solution on Intel platforms with gained accuracy.
Enhao Gong CEO & Founder, Subtle Medical
By leveraging the OpenVINO toolkit on 3rd Gen Intel Xeon Scalable Processors, the AI Builders team and Intel engineers helped us achieve 2.1x performance improvement on our latest NLP based application, which allows us to offer competitive deployment options on Intel Xeon CPU platform.
LIU MINGQIAN Director of Winning Health AI Lab
Thanks to the AI Builders program and the Project Team’s collaboration, Intel has strongly helped us in our journey, optimizing the training time of the NLP models of our product (Crystal), by nearly 20%.
Lorenzo Fornaciari VP of Strategy, iGenius
Intel hardware in combination with Intel optimization for TensorFlow has notably improved the performance of our algorithms, enabling not only faster training processes and inference on several neural network topologies, but also higher volumes of data to produce accurate and effective models.
Rafael San Miguel Carrasco ML OPS
I had the pleasure of co-presenting at the Intel Vision 2022 event to discuss how Intel and Caresyntax partnered to marry our AI/ML algorithms, such as for clip-assist and phase detection, with Intel’s high performance chip technology.
Ken Wu EVP & Chief Technology Officer, Caresyntax
The delivered improvements with OpenMP, found in Intel’s oneAPI Base Toolkit, bring important inference speedups that allow us to do real-time predictions to our Snyk Code users on 2nd and 3rd Gen Intel Xeon Scalable processors.
Veselin Raychev Director of AI, Snyk
We have worked closely with the Intel team to optimize the performance of some of our core algorithms used for our anomaly detection and time series forecasting models leveraging software libraries like IPP and oneDAL, found in Intel's oneAPI Toolkits, on Intel 3rd Gen Xeon Scalable processors.
Ira Cohen Co-Founder & Chief Data Scientist, Anodot
OpenVINO toolkit is seamlessly integrated with DeepEdge’s DL Ops platform for model conversion, optimization, and deployment. The platform also leverages other tools from the OpenVINO toolkit including post optimization toolkit and Neural Network Compression Framework.
Anup Mehta Co-Founder & CEO, DeepEdge
Working with Intel has been a pleasant experience. Our People Tracking (crowd behavior analytics) solution benefits from higher throughput by using the Intel OpenVINO Toolkit, and the scalability offered by the Intel 3rd Gen Xeon Scalable Processor.
Novan Parmonangan Simanjuntak Head of AI & ML Strategy, GLAIR.ai
Intel team helped us optimize our training time with Intel optimizations for TensorFlow such as oneDNN, and along with code optimization they have added modules which support multithreading and maximum CPU core utilization reducing the overall dependencies on GPUs.
Akshay Mohite Sr. Machine Learning Engineer, Climate Connect
The Intel team shared excellent insights and knowledge and we were able to experience up to a 2.7X inference throughput improvement using Intel Extension for PyTorch on Intel Xeon CPUs for our SimplyBot OCR engine.
Rohit Gupta CEO and Co-Founder of Simplify360
The insights shared by Intel team on how we can speed up our NLP pipeline by optimizing the ML models for faster inference have been significantly valuable to us.
Mahesh Maan Team Lead, Data Science Team, GreyB
iDeepwise has benefitted greatly from our collaboration with Intel. The AI Builders team has helped us achieve a 3.1x inference performance boost with our Medical Image Diagnosis Solution.
Xiwen Zhang Marketing Director, iDeepwise
Optimisation with OpenVINO leads to lower costs almost by 3 times, greater flexibility to operate in data centers worldwide where no GPUs are available.
Arjan Heijl Managing director Enbisys Netherlands
Based on AI Builders teams kind and professional support, we were able to achieve 3.25 times higher throughput inference time without compromising the 3D Hand Pose Estimation model accuracy.
Hanhim Cheng - CEO Co-Founder, EnerzAI
Brainpool's collaboration with Intel led to the development of an AIaaS platform which leverages an English RoBERTa NLP Sentiment Analyzer model which is optimized using OpenVINO
Dr. Peter Bebbington Chief Technology Officer, Brainpool
Recurrent AI extensively uses Intel technology-based computing platforms in AI model inference
Dr. Yutao Zhang Co-founder, CTO - Recurrent AI
The collaboration with Intel has been wonderful; they helped us optimize our code which improved our AI model's performance by more than 30%
Dr. Ranjan Satapathy Senior Data Scientist at Graphene AI
Thanks to oneDNN enabled TensorFlow on Intel 2nd Gen Xeon Scalable CPU systems (Cascade Lake), the optimization of D8AI’s proprietary NLP models were highly effective.
Dr. Richard Sheng GM, CTO & Co-Founder of D8AI
Intel® was able to optimize our AutoML function from a hardware perspective, using the full capacity of Intel’s hardware. Through Intel’s Xeon CPU, we were able to get up to a 200% increase in performance
Nelson Lee CSO of Synergies Intelligent Systems
We have been working closely with Intel to integrate OpenVINO to give our users the option to optimize their models before exporting/deploying, and the support from Intel has been world class.
Martin Isaksson CEO & Co-Founder at PerceptiLabs
We are thankful for the support we have received through the Intel AI Builders program which has helped us develop a streamlined process for taking models into production on Intel hardware and ensuring maximal performance is reached.
Nico Holmberg Lead AI Scientist, Silo AI
We got some great help from Intel AI Builders to optimize our solutions for Intel platforms. The main advantage I see with using Intel hardware is that we become less reliant on GPU’s.
Erik Landolsi CTO of Irisity
Accelerating workloads with Intel’s TensorFlow optimizations, we are able to help translators work faster with high accuracy.
Joern Wuebker Director of Research at Lilt
We had the support and valuable insights from Intel's engineers, giving us recommendations on how to make the best use of Intel hardware. We have obtained amazing performance times.
Mateus Dalponte Head of Data Science at Axur
By integrating the Intel AI stack with our s/w stack, we are able to rapidly deploy high performance serving in both development and production environments and are hoping to increase ROI for our clients.
Sathish Kumar EV Head of Product Management – XtractEdge from EdgeVerve
Working with the Intel AI Builders Program we got amazing support from the team and we were able to cut the inference time by half.
Pranabesh Dash Founder & CEO Xplorazzi
We at FootfallCam have a seamless collaboration with Intel AI Builders to improve AI application stack using Intel optimized OneDAL toolkit.
Vincent Chee FootfallCam Software Architect
We were fully dependent on GPUs for any model inference activities before we were introduced to OpenVINO platform and its capabilities. It is here that Intel teams have helped us a lot in optimising the Deep Learning models using their OpenVINO platform to infer the models in CPUs which was a great cost reduction in our projects.
Shivendra Jadon Senior Director, R&D, Qualitas Technologies
As technology evolves so fast, it’s especially important to stay at the forefront, and Intel makes it possible. We are excited to continue our collaboration for a better future.
Harry LiimalHead of Business Development, MINDTITAN
We are just so pleased that we have this partner (Intel) to work with us to make sure we are optimizing our capabilities to keep up with the need of the market and ultimately for us to protect our customers.
James CarnallManaging Director and VP of Services
The support we received from the Intel AI Builders team has been tremendously helpful. Magic Data is benefited in many ways such as Intel's cutting edge technical expertise, marketing opportunities and etc.
Helen WangDirector of Marketing – Magic Data
based on the advantages of Intel AI product portfolio, and the collaborative innovation of the software and hardware, the product performance will also be continuously optimized to the extreme.
Weibin ZhangCofounder, Chief Scientist - VoiceAI
Our membership in the Intel AI Builders program has helped us to develop and deploy these optimized solutions to the market, and to test on the next generation Intel chips and technology, which is great and totally exciting for us to get into the marketplace for our customers.
Soundar RajanHead of Virtual Workforce Automation, Perpetuuiti
we are glad that we are using Intel technologies from software as OpenVINO to hardware Intel Xeon CPUs. We start the cooperation using an optimization toolkit for neural network models that boost our inference of image recognition models by 3 to 4 times.
MICHAL LUKÁČ0Founder and Machine Learning Specialist , Ximilar
We’re very happy and proud about our growing relationship with Intel. We’re firm believers in standards and open source, particularly open source cryptography, because open standards and open source implementations are required to trust cryptographic libraries.
Dr. Kurt RohloffCTO and co‐founder of Duality Technologies
Leveraging AVX-512 on the latest Xeon Scalable processors provides a significant speed-up, and we look forward to being able to provide this to customers and to continuing to develop our relationship with Intel.
Helen KimVP of Wireless Technology and IP at MaxLinear
To date the gains resulting from this collaboration reveal enormous potential to further accelerate the calculation of the AI results, with a positive outcome for both clinicians and patients.
Jaap KroesResearch Team Lead at ScreenPoint Medical
we've been thrilled to partner with Intel in making COVID-Net CT available to front-line and hospital workers using Intel processors. Our collaboration has been one of the highlights during these challenging times.
Sheldon FernandezCEO at DarwinAI
We are honoured to be part of Intel’s AI ecosystem and for our close collaboration in helping telecom operators transition to cloud-native 5G network and achieve automated orchestration and operations. This partnership enables us in making our deep learning models for predicting network issues available to run efficiently on Intel platforms.
Rajendra PandaCo-founder, Avanseus.
Adopting Seassoon’s big data platform, enhanced with Intel software optimizations, Shanghai Pudong Government Service Center automated the screening, approval, and audit processes for 1000+ administrative tasks with improved transparency, efficiency, and customer services.
Jiangang DuanCo-founder and CTO of Seassoon
With the 3rd Generation Intel Xeon processors we see a great uptake in terms of performance.
Ariel PisetzkyVP of IT and Cyber at Taboola.
We very much appreciate the professional and helpful work done by the AI Builder team, and look forward to collaborations down the road that will lead to improved performance of our genomics tools.
Rafael AldanaProduct Application Director of Sentieon Inc.
Intel team has offered not only the technical consultation to help us optimize the solution but also the professional training of the latest Intel AI technologies in general.
LIU MingqianDirector of Winning Health AI Lab
From a pure R&D perspective it has been very valuable to be able to evaluate the hardware with support from Intel.
Stefan NilssonComputer Vision Specialist, Irisity
Intel AI Builders is definitely one of the most exciting programs, from which we are given access to intel’s hardware supporting the next generation of confidential computing
Lei ZhangDirector of Information Security, iExec
is that shifting from GPU based architectures to Intel Xeon and OpenVINO improved our speeds by several times more than we anticipated. So, it’s been a fantastic journey
Sharath PuranikDirector of Product & Engineering at Stylumia
We were able to learn how to optimize using Intel’s distribution of the OpenVINO toolkit, so that we can infer using a converted IR model. What that meant for us was a significant improvement increase over what we had earlier as a baseline
Shameer ThahaGlobal Chief Strategy Officer, Accubits
The AI Builders team has helped us a lot with benchmarking our translation system and helping to make it really efficient by putting in CPU specific optimizations
Spence GreenCEO of Lilt
By optimizing the AI models that run on Intel's hardware, Deci enables customers to get even more speed, allowing for cost-effective and more general deep learning use cases on Intel CPUs.
Yonatan GeifmanDeci CEO and co-founder
Accelerating model inference using OpenVINO on Intel Xeon CPUs was very insightful for us and we look forward to further collaboration.
Jonathan PirnayTechnical Consultant AI, Fujitsu Germany
Matroid prides ourselves on being able to help enterprises deploy computer vision detectors quickly and easily, on our customer's premises or in the cloud.
Reza ZadehCEO of Matroid & Adjunct Professor at Stanford University
John Snow Labs is excited to put the AI acceleration features of the 3rd Gen Intel Xeon Scalable processors to good use.
David TalbyCTO, John Snow Labs
With Intel’s collaboration and use of AVX512, we reached a 4 fold speedup in the calculation time. Using Intel’s newest chipset AVX512 we were able to speed up CPU-oriented tasks (preprocessing) and also matrix operations (CNNs).
Chris SchneeCMO, MaxQ AI
The incredible flexibility of new 3rd Gen Intel® Xeon® Scalable processors empowers us to select higher throughput versus lower latency by varying the number of streams making Intel technology ideal for batch processing and transactional workloads.
Jaya Kishore ReddyCTO & Co Founder
We strongly believe that AI Builders is the perfect vehicle for creating successful commercial AI solutions.
Ari AaltoVP Technology at TietoEVRY PDS
The resources and the expertise we get from the teams at Intel has really made us understand the power of software tools within the Intel ecosystem.
Anshuman PandeyFounder & CEO of NimbleBox
Our collaboration with Intel AI Builders has literally saved us years of development, in terms of human effort.
Vishal RaiCofounder and CEO of Embold
Thanks to this optimization for multiple use cases we have seen 2 to 3X improvement in performance while not compromising accuracy.
Ganesh IyerCTO of Knowledge Lens
With a high-performance medical application like ours, a major share of the AI workloads are carried out by the Intel® Distribution of OpenVINO™ toolkit.
Chris SchneeCMO, MaxQ AI
The co-operation we had with the Intel AI Builders team was flawless and very professional.
Michal LukáčFounder and Machine Learning Specialist, Ximilar
it's crucial to experiment and iterate quickly and this is something the Intel® AI Builders program enabled us to do.
Agrita GarnizoneHead of Data, Nordigen
The Intel® oneAPI AI Toolkit's PyTorch 1.6 with integrated Intel® oneAPI Deep Neural Network Library delivered up to 11.4X¹ faster inferencing for digital pathology medical screening.
GUI KUNVice President, KFBIO