Back to Programs

Development tools

OpenVINO Toolkit

Make AI inference faster and easier to deploy!

armassemblyc++pythonx86
Organization website

Participation history

5 GSoC years

20262025202420232022

2026

18 projects

Official year page

Synthetic Anomaly Generation

Unsupervised anomaly detection models in Anomalib rely on synthetic anomalies to calibrate decision thresholds, but the current Perlin-based pipeline...

Project page

AVX2-Accelerated Collision Detection for Real-Time MPC in Humanoid Robotics

This project aims to resolve the computational bottleneck in self-collision detection within the OCS2 framework for humanoid robotics. The current...

Project page

Demonstrating integration of openhands with OpenVINO Model Server

This proposal aims to create a production-ready, single-command deployment integrating the OpenHands AI coding agent with the OpenVINO Model Server...

Project page

OpenVINO GenAI: Add Image-to-Video Support to LTX Video Generation Pipeline

OpenVINO GenAI currently supports text-to-video generation via the LTX pipeline but lacks image conditioning. This project adds Image-to-Video (I2V)...

Project page

OpenVINO Profiling using VTune & GTPin

The existing GPU profiling workflow in OpenVino, through benchmark_app connects kernels to the model layers (primitives) and reports their execution...

Project page

Agentic AI for Predictive Maintenance using OpenVINO

This project aims to develop an agentic AI system for predictive maintenance of urban infrastructure using OpenVINO. The goal is to enable...

Project page

GGUF Reader V2 in OpenVINO GenAI for Direct GGUF Execution via Dynamic Graph Translation

The current GGUF reader in OpenVINO GenAI manually reconstructs model architectures by parsing metadata and building the OpenVINO model layer by...

Project page

Agentic Toolkit for AI PC with OpenVINO

OVAT is a modular Python toolkit with a CLI for building and running local AI agents on Intel AIPCs. Uses OVMS as the inference backbone, adds...

Project page

Optimize a VLA Model for Embodied Intelligence: OpenVINO Toolkit (Intel) Project #21

VLA models like unifolm-vla combine a vision-language backbone with a flow matching action head. On Intel iGPU, these models face two major...

Project page

Automated Agent System for Adding New Model Support to Optimum-Intel

Develop an autonomous multi-agent system using LangGraph to automatically analyze Hugging Face model architectures and generate the necessary code,...

Project page

No Code AI Workflow Automation with n8n and OpenVINO Model Server for GPU and NPU

This project integrates n8n, a no-code workflow automation platform, with OpenVINO Model Server to let users build AI-powered workflows that run on...

Project page

Web-Based AI-Assisted OS Image Builder for Intel Edge AI Suites

Deploying Intel Edge AI Suites (Metro AI, Manufacturing AI, Retail AI, Robotics AI) requires users to manually configure OS images with correct...

Project page

Multi-Object Tracking Integration for OpenVINO Training Extensions (OTX)

OTX supports 15+ detection models but has no built-in tracking.Every real-world video application (conveyor belt counting, asset monitoring,...

Project page

Optimizing Quantized Model Inference on ARM with OpenVINO

This project aims to improve INT8 inference performance for quantized YOLO26 models on ARM devices using OpenVINO. Currently, several transformation...

Project page

Continuous Face-Detection with Automatic Device Switching on AI PCs using OpenVINO AUTO feature

Modern AI PCs have multiple inference devices (CPU, GPU, NPU) but most applications use only one device at a time. This project aims to build a...

Project page

Auto-Labeling "Data Factory" & Edge Training Integration

This project introduces a ‘Cold Start’ Data Factory that turns zero/one-shot prompting into a full data-to-training workflow. The core contribution...

Project page

Build a GUI Agent with Local LLM/VLM using OpenVINO for Desktop Automation

A lot of everyday computer tasks—like navigating applications, managing files, and configuring settings—are repetitive and require multiple manual...

Project page

Object Tracking in MP with OpenVINO Inference

This project aims to integrate bytetrack algorithm into a Mediapipe graph using OpenVINO inference calculators. The bytetrack graph will be deployed...

Project page

2025

14 projects

Official year page

Interactive Multimodal Embedding Explorer

Multimodal Embedding Explorer (MMEE) is an interactive platform for exploring, cleaning, and validating large-scale multimodal datasets combining...

CodeProject page

OpenVINO GenAI provider for LangChain

This project aims to develop an OpenVINO GenAI extension for LangChain, enabling users to utilize OpenVINO as a backend for running language models....

CodeProject page

Enhancing Anomalib with SOTA Anomaly Detection Benchmarks, Models and Evaluation Metrics

This project aims to enhance Anomalib by integrating state-of-the-art (SOTA) anomaly detection models, including GLASS, GLAD, MuSc, for both...

CodeProject page

Improve OTX Classification via DoRA and Transformer Backbone

This project addresses the inefficiency of full fine-tuning in classification tasks on small-scale datasets and low-resource environments within...

CodeProject page

Optimized Serving endpoint for classic machine learning model training and inference

As machine learning models grow in size and complexity, there is a rising need for fast, efficient, and scalable ways to serve them in production...

CodeProject page

Integration of Open WebUI with OpenVINO Model Server

Open WebUI is a very popular component that provides a user interface to generative models. It supports use cases related to text generation, RAG,...

CodeProject page

Traffic Intersection Monitoring System with OpenVINO

This project develops an AI-powered desktop application for real-time traffic monitoring and intelligent automation. It detects vehicles,...

CodeProject page

Fine-tuning Vision-Language Models for Object Detection and Hierarchical Classification

This project aims to fine-tune Vision-Language Models (VLMs) for domain-specific tasks like object detection and hierarchical classification,...

CodeProject page

OpenVINO AI PC Model Training Kit

This proposal outlines a detailed, structured, and actionable two-phase strategy for developing an OpenVINO AI PC Model Training Kit. The core...

CodeProject page

Keras hub GenAI/LLM pipelines for the OpenVINO backend in Keras 3 with Export support

Enable popular Keras Hub GenAI/LLM pipelines for the OpenVINO backend in Keras 3 and optimize their performance. This project focuses on implementing...

CodeProject page

Project 18: Multimodal embeddings based on OVMS

This project aims to optimize the calculation of multimodal embeddings (text, image, audio, video) by exposing REST API endpoints in OpenVINO Model...

CodeProject page

OpenVINO Messenger AI-Assistant for an AI PC

Modern messaging platforms like Discord and Slack are not just used for communication but also for consuming news, discussions, and topic-specific...

CodeProject page

Gesture Control with OpenVINO

This project aims to develop a hand-gesture computer control system by porting MediaPipe’s Gesture Recognizer to OpenVINO, enhancing performance on...

CodeProject page

Training Edge-Deployable Models via Collaborative Annotation with Humans and Large Foundation Models

This project aims to implement a framework that leverages Large Foundation Models (LFMs) for automated dataset annotation. It also aims to distill...

CodeProject page

2024

8 projects

Official year page

Porting and Optimizing OpenVINO for RISC-V Devices with Vector Extension

[Problem] The project aims to enhance the runtime performance of OpenVINO on RISC-V devices to fully leverage their power efficiency for running...

CodeProject page

OpenVINO adapters - PyTorch adapter

This project aims to make PyTorch and OpenVINO interchangeable. By using the Python API of OpenVINO, we will create an adapter for PyTorch, the first...

CodeProject page

1. OpenVINO Extension for Automatic1111 Stable Diffusion WebUI

This proposal is provided by OpenVINO team. I just copy and paste here. Short description: Automatic1111 is a powerful web user interface based on...

CodeProject page

VisionGuard: Smart Screen Time Management

VisionGuard is a desktop application designed for AI-enabled PCs to mitigate eye strain through real-time gaze tracking. The app builds on OpenVINO's...

CodeProject page

Inference of Neural model in Node.js environment using OpenVINO

This project targets the expansion of OpenVINO's functionality within the Node.js ecosystem, the Node.js API which currently lacks comprehensive...

CodeProject page

OpenVINO GenAI

I will extend the existing OpenVINO GenAI text-to-image pipeline to support image-to-image functionality by incorporating ControlNet and utilizing...

CodeProject page

Improve Generative AI workload performance on ARM devices with OpenVINO

This project aims to enhance the performance of Generative AI (GenAI) workloads on ARM devices using the OpenVINO toolkit, focusing on reducing...

CodeProject page

AnomalibVLM: Enhancing Anomaly Detection with Vision Language Models.

Problem: Traditional anomaly detection systems often struggle with limited data scenarios and can lack the ability to explain their detections....

CodeProject page

2023

5 projects

Official year page

Industrial meter reading with OpenVINO

The aim of the project is to read industrial meter with Deep Learning frameworks and OpenVINO .The project is divided into two parts. In the first...

CodeProject page

Add ensembling methods for tiling to Anomalib

When tackling the challenge of detecting defects in high-resolution images, we encounter many challenges. One significant problem is the difficulty...

CodeProject page

Anomaly Segmentation Metrics for anomalib

This project proposes novel evaluation metrics for anomaly segmentation in computer vision, taking into account pixel-level and spatial information....

CodeProject page

Add OpenVINO support to John Snow Labs Spark NLP

Performance-focused, production-level machine learning libraries need to leverage the resources at their disposal to the maximum extent to deliver...

CodeProject page

Showcase performance of PyTorch Image Models (Timm) with OpenVINO

My proposal aims to optimize PyTorch Image Models (Timm) using OpenVINO and NNCF libraries. The problem at hand is that these models are...

CodeProject page

2022

3 projects

Official year page

3D Jupyter Notebooks and WebGL Integration for OpenVINO

We usually need visualization, in fact, to enhance the interpretability of data or results, and facilitate the understanding of developers and users....

CodeProject page

Object Detection demo application to run AI on ARM-based Android devices

The openvino_contrib contains The ARM CPU Plugin project, which provides support for ARM CPUs, and Java Wrappers access The underlying infering...

CodeProject page

Train a DL model for synthetic data generation for model optimization

The project is divided into two parts. The goal for the first part is to train a lightweight Deep Learning model to generate a dataset of synthetic...

CodeProject page