MVTec AI Vision Solver is bringing AI agents into machine vision development, with MVTec set to demonstrate a prototype at VISION 2026 that can turn natural-language descriptions of machine vision tasks into HALCON-based applications.
MVTec is taking its use of artificial intelligence in machine vision into a new part of the development process.
At VISION 2026 in Stuttgart, the company will demonstrate its AI Vision Solver, a prototype designed to use AI agents to help developers create machine vision applications. Instead of using AI only to analyze images, MVTec is exploring how AI can assist with deciding how an image processing application should be built.
The prototype combines MVTec HALCON, the company’s machine vision software platform, with AI agents and the new HDevelopEVO integrated development environment.
Visitors to VISION 2026 will be able to see the sysem in operation at MVTec’s booth 8C56 in Hall 8 from October 6 to 8, 2026.
MVTec Moves AI Into Machine Vision Development
MVTec has been using deep learning in machine vision since 2016. Its AI technologies are already used for applications such as classification, inspection and object recognition.
The AI Vision Solver takes a different approach.
A developer can describe a machine vision task using natural language and images. The AI agents then interpret the requirements and determine an appropriate image processing workflow.
The system can select operators from the HALCON library and generate an executable machine vision application inside HDevelopEVO.
Dr. Maximilian Lückenhaus, Director of Marketing + Business Development at MVTec, explained that the company is integrating its software with AI agents to make HALCON easier to use while reducing development effort.
The distinction is important. MVTec is not presenting AI as another inspection algorithm. The AI is being introduced at the application development stage.
AI Agents Use MVTec’s Machine Vision Knowledge
The AI Vision Solver is also designed around a predefined knowledge base supplied by MVTec.
That gives the AI agents access to MVTec’s machine vision expertise when determining how a particular task should be implemented.
For industrial users, this is an important part of the concept.
A general AI model can generate code or suggest approaches, but machine vision development requires knowledge of image processing operators, application requirements and the relationship between different processing steps.
MVTec’s approach is to combine AI agents with a defined machine vision software environment and domain-specific knowledge.
The result is intended to help the AI make more relevant decisions when creating an application.
From Natural Language to a Working Vision Application
At VISION 2026, MVTec will demonstrate the system using practical machine vision tasks.
Visitors will be able to select applications such as:
- Barcode reading
- Pattern matching
- Optical character recognition (OCR)
The task can then be provided to the system through prompts.
The AI agents interpret the request, select the required HALCON operators and generate the application in HDevelopEVO.
The system can also identify syntactic errors and logical gaps in the generated application. That information is returned to the AI agents, allowing them to refine the result.
This creates a development loop in which the AI is not simply generating an initial answer. It can also participate in the process of identifying and correcting problems.
For machine vision developers, that could reduce the amount of repetitive programming required when creating an application.
HDevelopEVO Provides the Development Environment
The AI Vision Solver is built around HDevelopEVO, MVTec’s next-generation integrated development environment for HALCON.
HDevelopEVO is intended to provide a more structured and transparent environment for machine vision development. MVTec plans to gradually replace its current IDE with HDevelopEVO starting in 2027.
That makes the new development environment an important part of the company’s AI strategy.
AI-assisted application development requires an environment where generated workflows can be created, inspected, tested and modified. HDevelopEVO provides that environment for the AI Vision Solver.
VISION 2026 will therefore give visitors not only a demonstration of the AI prototype, but also an early look at the development environment that will underpin MVTec’s next generation of HALCON development.
AI Can Assist, But Engineers Still Make the Decisions
MVTec sees the AI Vision Solver as a way of reducing development time rather than removing the developer from the process.
Instead of manually programming every individual processing step, developers could increasingly focus on defining the task, evaluating the generated workflow and optimizing the final application.
Lückenhaus emphasized that AI agents can reduce development effort, but the technology does not take responsibility for the industrial application.
The final assessment remains with the user.
That is particularly relevant in manufacturing, where a machine vision application may be used for quality inspection, identification, measurement or other production-critical tasks.
The ability to generate an application more quickly does not remove the need to verify that the application delivers the required performance under real production conditions.
MTN Analysis: AI Is Moving Upstream in Machine Vision
The most interesting part of MVTec’s announcement is not that the company is using AI in machine vision.
AI has already become an established technology for many machine vision applications.
The more important development is where MVTec is applying AI.
Traditional AI-based machine vision focuses on what happens after an image has been captured. Deep learning models can classify objects, detect defects or recognize patterns.
The AI Vision Solver addresses what happens before that point. It helps determine how the machine vision application should process the image in the first place.
That represents a move from AI-powered vision to AI-assisted vision engineering.
For experienced developers, the potential benefit is a reduction in repetitive development work. Instead of manually selecting and connecting every processing step, the developer can describe the intended task and use AI to generate a starting point.
For less experienced users, natural language interaction could also make sophisticated machine vision software more accessible.
However, the quality of the final application will still depend on validation and engineering knowledge. A generated workflow can be syntactically correct and still be unsuitable for a particular production environment.
This is why MVTec’s decision to keep the developer responsible for the final application is significant.
The AI Vision Solver is better understood as an engineering assistant than an autonomous machine vision system.
The combination of AI agents, HALCON’s established operator library and MVTec’s domain-specific knowledge could also point toward a broader direction for industrial software. AI assistants are increasingly moving inside engineering platforms, where they can work with the tools, libraries and constraints that engineers already use.
For machine vision, that could eventually change how applications are specified and developed.
MVTec Brings More AI Technology to VISION 2026
The AI Vision Solver will be one of the main highlights of MVTec’s VISION 2026 presence, but the company will also showcase its broader machine vision portfolio.
The exhibition will include new releases of:
MVTec will also demonstrate technologies aimed at accelerating AI algorithms through the integration of high-performance hardware accelerators.
Together, the technologies demonstrate how AI is becoming integrated across multiple stages of machine vision, from image analysis and deep learning to application development and hardware acceleration.
Where to See MVTec at VISION 2026
MVTec will exhibit at VISION 2026 in Stuttgart from October 6 to 8, 2026.
Hall 8, Booth 8C56
Visitors can see the AI Vision Solver prototype in operation and get an early look at HDevelopEVO.
FAQ
What is the MVTec AI Vision Solver?
The AI Vision Solver is a prototype that uses AI agents to help developers create machine vision applications. Users can describe a task using natural language and images, with the system then generating a corresponding HALCON-based workflow.
How does the AI Vision Solver use AI agents?
The AI agents interpret the user’s requirements, select suitable HALCON operators and generate an executable machine vision application. They can also use feedback about syntax errors and logical problems to refine the generated application.
Does the AI Vision Solver replace machine vision developers?
No. MVTec describes the technology as a development aid. Developers remain responsible for evaluating, validating and optimizing the resulting machine vision application.
What is HDevelopEVO?
HDevelopEVO is MVTec’s next-generation integrated development environment for HALCON. It is designed to provide a more structured and transparent environment for machine vision application development.
When will HDevelopEVO replace the current MVTec IDE?
MVTec plans to gradually replace its current IDE with HDevelopEVO starting in 2027.
How is the AI Vision Solver different from traditional machine vision AI?
Traditional machine vision AI can be used to classify, inspect or recognize objects and patterns in images. The AI Vision Solver focuses on the development stage, using AI agents to help determine and generate the image processing workflow itself.
Where can I see the AI Vision Solver?
MVTec will demonstrate the AI Vision Solver at VISION 2026 in Stuttgart from October 6 to 8, 2026, in Hall 8, Booth 8C56.
Further Reading
MVTec MERLIC AI Machine Vision Moves From Endress+Hauser Pilot to Global Assembly Rollout
For more information about MVTec’s AI and machine vision technologies, visit the official MVTec website.
