STOCK TITAN

Cognex Launches In-Sight Vision Controller Powered by NVIDIA

Rhea-AI Impact
(Moderate)
Rhea-AI Sentiment
(Neutral)
Tags

Cognex (NASDAQ: CGNX) launched the In-Sight 6900 Vision Controller, a modular AI vision controller powered by NVIDIA Jetson for high-capacity edge AI processing.

Key features: interchangeable cameras/optics/lighting, transformer-based few-sample classification (10–20 images), pixel-level segmentation, integration with OneVision, and up to 157 TOPS performance with NVIDIA TensorRT for real-time inference on high-speed lines.

Loading...
Loading translation...

Positive

  • Modular vision architecture supports interchangeable cameras, lenses, and lighting
  • Few-sample classification with transformer models requiring 10–20 training images
  • Pixel-level segmentation for robust feature isolation on challenging surfaces
  • Integration with OneVision for centralized development and global deployment
  • Up to 157 TOPS AI performance via NVIDIA Jetson enabling parallel high-resolution models
  • Real-time inference optimized with NVIDIA TensorRT for microsecond-level production timing

Negative

  • None.

Key Figures

Share price: $55.09 Daily move: 1.53% 52-week range: $25.70–$59.88 +5 more
8 metrics
Share price $55.09 Pre-news price on publication day
Daily move 1.53% Price change over prior 24 hours
52-week range $25.70–$59.88 52-week low and high pre-news
2025 revenue growth 9% Year-over-year revenue increase in 2025
Operating margin improvement 12.6% to 16.3% Operating margin change in 2025
Capital returned $206 million Dividends and buybacks in 2025
AI performance 157 TOPS Max AI throughput of In-Sight 6900 Vision Controller
Training images 10–20 images Few-sample classification requirement per model

Market Reality Check

Price: $55.09 Vol: Volume 2,337,779 is 46% a...
normal vol
$55.09 Last Close
Volume Volume 2,337,779 is 46% above the 20-day average of 1,602,833, indicating elevated interest ahead of and around the AI product launch. normal
Technical Shares at $55.09 are trading above the 200-day MA of $43.8 and about 8% below the 52-week high of $59.88.

Peers on Argus

CGNX gained 1.53% while momentum-flagged peers COHR and ITRI fell about 5.1–5.3%...
2 Down

CGNX gained 1.53% while momentum-flagged peers COHR and ITRI fell about 5.1–5.3%. Broader sector scanners show downside pressure, but CGNX’s positive move diverges from peers’ downside, pointing to company-specific strength around the AI vision controller launch.

Historical Context

5 past events · Latest: Apr 16 (Neutral)
Pattern 5 events
Date Event Sentiment Move Catalyst
Apr 16 Earnings call timing Neutral +0.6% Set dates for Q1 2026 earnings release and conference call.
Apr 06 Business divestiture Neutral +4.4% Closed sale of Japan-focused trading unit viewed as non-material to operations.
Mar 23 AI adoption study Positive +3.2% Released survey showing strong and growing AI vision adoption across manufacturers.
Feb 18 Conference presentation Neutral -0.7% Announced CEO appearance at Raymond James institutional investor conference.
Feb 17 Board refresh Positive -2.9% Added two directors with decades of experience to support AI strategy.
Pattern Detected

Recent CGNX news events skew positive-to-neutral and have more often been followed by modestly positive price reactions, with only the board refresh drawing a negative move against constructive strategic messaging.

Recent Company History

Over the last six months, Cognex news has centered on portfolio reshaping, AI positioning, and governance. A Feb 17 board refresh supported its AI vision focus but drew a -2.89% move. A Mar 23 AI adoption study saw shares rise 3.17%. The Apr 6 Japan divestiture closed at about $11.9 million and was followed by a 4.38% gain. An upcoming Q1 2026 earnings release on May 6 is the next scheduled catalyst context for today’s AI product launch.

Market Pulse Summary

This announcement introduces Cognex’s In-Sight 6900 Vision Controller, pairing modular hardware with...
Analysis

This announcement introduces Cognex’s In-Sight 6900 Vision Controller, pairing modular hardware with NVIDIA-powered edge AI, including up to 157 TOPS of performance and few-sample training using 10–20 images. It builds on a 2025 backdrop of 9% revenue growth and operating margin rising to 16.3%. Investors may watch upcoming Q1 2026 earnings on May 6 for early signs of commercial uptake and how this product fits within Cognex’s broader AI vision strategy.

Key Terms

neural networks, segmentation
2 terms
neural networks technical
"high-performance AI engine capable of executing sophisticated neural networks directly at the edge."
A neural network is a type of computer model that learns to recognize patterns by adjusting many simple connections, similar to how a brain’s network of neurons strengthens certain pathways after practice. Investors care because these models can analyze large, messy datasets to forecast trends, detect fraud, or automate decisions, potentially improving trading signals or operational efficiency; their performance and transparency can materially affect a company’s competitive edge and risk profile.
segmentation technical
"Robust Segmentation: Pixel-level segmentation that isolates critical features, delivering reliable"
Segmentation is the practice of dividing a larger whole into smaller, more manageable parts—such as splitting customers into groups with similar needs, separating a company’s operations into reporting units, or isolating regions in a medical image. For investors, segmentation reveals where revenue, risk, or growth are concentrated, like reading a map that shows which neighborhoods drive traffic; that clarity helps assess performance, strategy and future prospects.

AI-generated analysis. Not financial advice.

Advanced AI Vision Controller delivers the modularity required to solve demanding, compute-intensive manufacturing applications without external PCs

NATICK, Mass., April 28, 2026 /PRNewswire/ -- Cognex Corporation (NASDAQ: CGNX), the global technology leader in industrial machine vision, today announced the launch of the In-Sight® 6900 Vision Controller. The modular In-Sight Vision Controller gives manufacturers the freedom to configure cameras, optics, and lighting precisely for their inspection requirements, eliminating the compromises of fixed-configuration systems. Powered by NVIDIA Jetson technology, it delivers high-capacity AI processing at the edge, enabling the most demanding industrial workloads to run without external PCs or complex distributed architectures. 

"As manufacturers accelerate their adoption of AI, they want solutions that are both powerful and easy to deploy," said Matt Moschner, President and CEO of Cognex. "The In-Sight Vision Controller delivers exactly that, combining NVIDIA's edge AI processing with Cognex's modular hardware and proven vision tools. This allows more customers to solve demanding inspection applications at the edge, without the cost and complexity of PC-based systems, and opens up applications that weren't possible before."

Advanced AI Tool Modes for the Toughest Vision Challenges
The In-Sight Vision Controller introduces Cognex's most advanced AI tool modes to date—purpose-built for challenging inspections, capable of handling parts of varying sizes, detecting highly variable defects, and delivering consistent results even in dynamic and unpredictable production environments.  

Key capabilities include:

  • Modular Vision Architecture: Support for interchangeable cameras, lenses, and industrial lighting enables precise system configuration, including Cognex's lenses, optics, cameras, and accessories (LOCA) portfolio.
  • Few Sample Classification: Transformer-based classification models requiring as few as 10–20 training images, reducing data collection time and accelerating deployment.
  • Robust Segmentation: Pixel-level segmentation that isolates critical features, delivering reliable analysis on challenging surfaces.  

The In-Sight Vision Controller also integrates seamlessly with OneVision™, Cognex's collaborative AI platform that allows engineering teams to develop, refine, and scale inspection applications globally from a single environment. Manufacturers can roll out improvements faster and maintain consistent quality across plants, production lines, and shifts. 

Powered by NVIDIA Jetson for Real-Time, High-Throughput AI at the Edge
By incorporating the NVIDIA Jetson module, the In-Sight Vision Controller transforms from a traditional image processor into a high-performance AI engine capable of executing sophisticated neural networks directly at the edge.  

Performance highlights include: 

  • Parallel Processing for Complex Inspections: Up to 157 TOPS of AI performance allows multiple high-resolution AI models to run simultaneously.
  • Real-Time Inference: Optimized NVIDIA TensorRT integration helps ensure AI decision-making stays synchronized with the microsecond-level timing of high-speed production lines. 

About Cognex Corporation
For over 40 years, Cognex has been making advanced machine vision easy, paving the way for manufacturing and distribution companies to become faster, smarter, and more efficient through automation. Innovative technology in our vision sensors and systems solves critical manufacturing and distribution challenges, providing unparalleled performance for industries from automotive to consumer electronics to packaged goods. Cognex makes these tools more capable and easier to deploy thanks to a longstanding focus on AI, helping factories and warehouses improve quality and maximize efficiency without needing highly technical expertise. We are headquartered near Boston, USA, with locations in over 30 countries and more than 30,000 customers worldwide. Learn more at cognex.com. 

Media Contact:
Liz Bradley – Head of Communications
Cognex Corporation
pr@cognex.com

IR Contact:
Greer Aviv – Head of Investor Relations
Cognex Corporation
ir@cognex.com

Cision View original content to download multimedia:https://www.prnewswire.com/news-releases/cognex-launches-in-sight-vision-controller-powered-by-nvidia-302753259.html

SOURCE Cognex Corporation

FAQ

What is the In-Sight 6900 Vision Controller announced by Cognex (CGNX)?

The In-Sight 6900 is a modular AI vision controller for industrial inspection at the edge. According to Cognex, it supports interchangeable cameras, optics, lighting, and runs high-capacity AI without external PCs.

How many training images does the In-Sight 6900 require for classification tasks (CGNX)?

Classification can work with as few as 10–20 images for some models. According to Cognex, transformer-based few-sample models reduce data collection and speed deployment.

What AI performance does the Cognex In-Sight Vision Controller (CGNX) deliver?

The controller delivers up to 157 TOPS of AI performance using NVIDIA Jetson. According to Cognex, this enables multiple high-resolution AI models to run in parallel at the edge.

Can the In-Sight 6900 run real-time inference on high-speed production lines (CGNX)?

Yes — it supports real-time inference synchronized to microsecond-level timing. According to Cognex, NVIDIA TensorRT optimization helps maintain decision-making on high-throughput lines.

How does the In-Sight 6900 integrate with Cognex software for deployment (CGNX)?

The controller integrates with OneVision for collaborative development and scaling across sites. According to Cognex, teams can develop, refine, and deploy inspection applications globally from one environment.