Merwansky·Nov 4TensorRT or OpenVINO? One Vision Model, NVIDIA and Intel Hardware, MeasuredComputer Vision & AI
Maks Khomutskyi·Oct 20Illama 2: Local LLMs on Intel ArcA practical OpenVINO setup, tested on an Arc Pro B50 with 16 GB VRAM.
El Psi·Oct 13Optimizing Multi-Model CPU Inference with OpenVINO: Concurrency, Threading, and SchedulingWhen optimizing machine learning inference, we usually focus on the models themselves: quantization, pruning, or choosing smaller…
InOpenVINO-toolkitbyPassavee Losripat·Oct 6GSoC 2026: Optimizing Quantized Model Inference Performance on ARM Devices with OpenVINOIn the AI era, we are building models that are bigger, smarter, and more capable than ever. But a powerful model is only half of the story…
Luís Condados·Oct 63D Vehicle Detection on a Laptop iGPUUrbanOmniDetect-2 draws 3D boxes and a top-down map from one uncalibrated camera. Quantized to INT8 with OpenVINO, it runs at 30 fps on a…
MORTEZA HEIDARI·Sep 30Stop Paying Per Token: Run a 27B Coding Model Locally in VS Code Copilot on Your Intel AI PCHow I turned Qwen3.8–27B into a free, private model in the VS Code Copilot model picker, running on the integrated GPU of an Intel Core…
Luís Condados·Sep 24YOLO-NAS Without Super Gradients: 201 FPS on a Laptop CPU, 0.4 AP Lost to INT8Originally published at https://condados.ai/blog/yolo-nas-without-super-gradients
Rippertnt·Sep 22Priced Out of NVIDIA? We Benchmarked Five LLM Engines on Intel’s 32 GB Arc Pro B70The card wasn’t the bottleneck. Five serving engines on identical silicon came out 3.9× apart — and running it five times found bugs in our…
InOpenVINO-toolkitbyVishwa Teja·Sep 14GSoC 2026: Implementation of the ByteTrack algorithm in MediaPipe with OpenVINO™ inference…
InOpenVINO-toolkitbyJohyeongseob·Sep 11Building an Agentic AI Predictive Maintenance Pipeline with OpenVINO™ and Intel® Metro AI SuiteCritical infrastructure (pipelines, utilities, and assets) requires continuous inspection to prevent failures, leaks, and safety incidents…
