Introducing Open Edge Platform 2025.1 — Enhanced AI Tools & Industry-Focused Applications at the Edge #410
biapalmeiro
started this conversation in
Show and tell
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hi everyone 👋,
We’re thrilled to announce the Open Edge Platform 2025.1 release, packed with exciting enhancements and industry-specific sample applications designed to accelerate your edge AI development. Below is a quick overview of the highlights—let's discuss!
Key Highlights:
1. Industry Sample Applications
Metro AI Suite – Smart Traffic Intersection
Real-time multi-camera analytics + sensor fusion (LiDAR/radar) for advanced traffic management (Video, Spatial AI).
Gen AI Capabilities
Manufacturing AI Suite – Wind Turbine Predictive Maintenance
Handles time series ingestion and anomaly detection—integrated with MQTT, OPC-UA, OPC data streams.
Retail AI Suite
2. Feature Enhancements
Geti™: Supports fine-tuning CV models on Intel Arc GPUs. Adds pose estimation, configurable model sizing, and optimized data augmentation.
DL Streamer Framework: Now Windows-compatible for fast media analytics at the edge.
DL Streamer Pipeline Server: Scalable microservice for up to 4 cameras—real-time counting, tracking, and detection.
OpenVINO™ Upgrades:
GGUF reader integration (support for llama.cpp models)
Optimized LoRA adapters (LLM, VLM, text-to-image pipelines)
Support for image-to-image and inpainting transformer pipelines (Flux.1, Stable Diffusion 3)
3. Platform Enhancements: Streamlining Remote Management & Resource Utilization
The 2025.1 release takes a major step forward in simplifying remote management and optimizing resource utilization for edge environments.
This combination of near zero-touch onboarding, hardware-level optimization, and secure scalability makes the Open Edge Platform not just a toolkit for building applications but a complete solution for running them optimally on Intel silicon.
Beta Was this translation helpful? Give feedback.
All reactions