Deployment

22 May 2026 • 6 min read

Gemini 3.5 Flash for Vision: Evaluation and Benchmarks

Google released Gemini 3.5 Flash May 19, 2026 at Google I/O and it is the highest performing frontier model on the Roboflow Vision Evals leaderboard, beating Gemini 3.1 Pro on counting and spatial reasoning while running about three times faster. Six months after Gemini 3 Pro set

21 May 2026 • 4 min read

Integrate Roboflow with OPC UA

This guide shows how to push Roboflow Workflow detections to an OPC UA server, making them available as live tags to dashboards, PLCs, and SCADA platforms alongside any other plant sensor. Prerequisites * Roboflow Enterprise plan — the OPC UA Writer Sink is enterprise-only * Docker Desktop * An existing OPC UA server — Ignition,

20 May 2026 • 4 min read

Roboflow MCP: Build Vision Apps with Claude, Codex, and More

The new Roboflow MCP Server connects Claude, Codex, and other AI agents to Roboflow so they can label visual data, train custom models, and build deployable vision pipelines.

20 May 2026 • 2 min read

Launch: 300+ OpenRouter models available in Roboflow

Starting today, Roboflow's integration of OpenRouter gives you access to over 300 models, including every major VLM plus specialized and open-source options, all from a single interface. Pick the right model for the job today, and change your mind later without rewriting your pipeline. A common pattern in

13 May 2026 • 5 min read

GPT-5.5: Vision Benchmarks & Use Cases

Explore OpenAI GPT 5.5 vision benchmarks. Learn how its new high-resolution encoder improves document parsing and how to use it in Roboflow Workflows.

20 Apr 2026 • 3 min read

Vision Events: Turning Images into Insights

Transform raw model predictions into trackable insights. Vision Events provides a secure, searchable record of your model deployments alongside images and custom metadata.

16 Apr 2026 • 4 min read

Serverless GPU Inference Cost Comparison: Roboflow, GCP, AWS, Azure

In this blog post, we will explore how different cloud providers compare in running custom vision model inference. Custom RF-DETR Model If you need OCR, face detection/comparison, or general-purpose COCO-20 classification and detection, the major clouds already offer plenty of managed options, such as AWS Rekognition, GCP Vision, and

26 Mar 2026 • 18 min read

How to Use Roboflow for Video-Heavy Pipelines

Learn to build high-throughput video pipelines with ByteTrack, Time-in-Zone logic, and Roboflow Workflows for cloud or edge.

19 Mar 2026 • 14 min read

How to Deploy Computer Vision

In this definitive guide, we break down the decisions required to build a production-grade vision system.

2 Mar 2026 • 5 min read

Inference as a Service: How Roboflow Makes Vision AI Production-Ready

This guide explores how to abstract away the complexities of GPU orchestration and hardware allocation. Roboflow offers a production-grade API with built-in active learning, model chaining, and auto-scaling to turn your vision models into real-world solutions.

3 Feb 2026 • 4 min read

Launch: Process data directly from Wasabi, Backblaze B2, Cloudflare R2, and more

We’ve expanded Batch Processing to support the wider ecosystem of S3-compatible storage. Learn how to run inference on terabytes of data stored in Wasabi, Backblaze, and Cloudflare R2, and more.

22 Dec 2025 • 4 min read

AI for Live Video: Introducing the Serverless Streaming API

Today we are announcing the Serverless Video Streaming API, the simplest way to run AI on live video streams. Previously, using AI to understand live video streams meant managing complex cloud infrastructure or configuring local hardware. The Serverless Video Streaming API removes those barriers, allowing you to run powerful vision

9 Dec 2025 • 4 min read

Introducing Roboflow Rapid: Text prompt to vision model in minutes

For most of computer vision's history, training has been extremely challenging. Going from an idea to a deployed application meant spending weeks labeling hundreds or thousands of images by hand just to see if a concept would work. Now the era of manual labeling is ending. Today, we

8 Aug 2025 • 4 min read

How to use YOLOE for Zero-Shot Object Detection & Segmentation

In computer vision, few model families have made as big an impact as YOLO. Originally designed for fast and accurate object detection, the YOLO architecture has evolved rapidly, now supporting both complex detection and segmentation tasks. With the rise of YOLOE and other enhanced variants, it's now easier

25 Jul 2025 • 9 min read

How to Build an iOS App with Visual AI Capabilities

In this guide, we’ll walk through using a custom object detection model tailored to identify glasses and show you how to seamlessly integrate it into an iOS app for instant live detection — all with minimal latency and maximum convenience. 0:00 /0:13 1× Millions of people rely on

3 Jul 2025 • 21 min read

How to Control a PTZ Camera with Computer Vision and Roboflow Workflows

Learn how to control a PTZ camera in a computer vision application.

26 Jun 2025 • 5 min read

How to Use Perception Encoder

Learn how to use and deploy Meta AI's Perception Encoder model with Roboflow Inference.

26 Jun 2025 • 6 min read

How to Deploy RF-DETR to an NVIDIA Jetson

Learn how to deploy RF-DETR to a Roboflow Jetson.

24 Jun 2025 • 8 min read

Launch: Roboflow Instant Models

Roboflow Instant is a free, quick-to-train, few-shot model that requires as few as a half-dozen labelled images to produce an accurate object detection model.

12 Jun 2025 • 4 min read

Launch: Manage Computer Vision Deployments with Device Manager

Learn how to use the Roboflow Device Manager to manage computer vision deployments.

28 Mar 2025 • 7 min read

Launch: Train and Deploy RF-DETR Models with Roboflow

Learn how to train RF-DETR models with Roboflow Train and deploy RF-DETR models with Roboflow Workflows and Inference.

13 Mar 2025 • 7 min read

Launch: Roboflow Batch Processing

Learn how to use Batch Processing to run multi-step vision AI workflows on a folder of images or videos.

27 Feb 2025 • 5 min read

What Is Edge AI And How Does It Work?

By running models "at the edge," companies can benefit from high performance, fast decision making, reduced cloud costs, and enhanced security.

2 Jan 2025 • 6 min read

Rock, Paper, Scissors with AI: How to Make Multiplayer Games

This is a template for making multiplayer games - such as rock paper scissors- that involve your hands and body using AI or computer vision. You can even submit new games to the repo and I will host them at https://handland.lol Included Multiplayer Games: The repo currently comes