VisionLabs Luna Line

Main article: Training artificial intelligence

2025: Luna Line Platform Introduction

VisionLabs on June 11, 2025 introduced a platform for creating AI models without writing code.

The Luna Line platform allows you to run and train computer vision models without writing code. This reduces the timing of the creation of neural networks and the cost of projects by 20-30%.

When introducing solutions based on computer vision, it is important for a business to quickly check hypotheses and constantly further study neural networks on new data. At the same time, developing an AI model from scratch is a long and expensive process. On average, the launch of the project takes 4-6 months and requires a team of ML specialists. With VisionLabs' Luna Line solution, the first version of the AI model can be created in 2-3 days and new versions can be launched just as quickly after further training.

Luna Line allows you to train ML models on just a few dozen images. The user needs to upload at least 50 pictures to the system and mark them up - add tags that will help the model understand what is in the image. Next, the user clicks "Train" - this stage can take from a few minutes to hours, depending on the amount of data for training. After that, the neural network is ready to work.

The tool will also be useful for experienced ML specialists: they can automate simple tasks and speed up product development. Visual language models (VLM, Visual Language Model) are also available on the platform, which are capable of working with both images and texts. They allow you to automatically mark up media files and analyze what is depicted on them.

The platform is available in the cloud, it can also be deployed on its own servers. The finished system allows you to work with both data from CCTV cameras and photos from your phone. For example, an online delivery service using Luna Line can automate product quality assessment: a chatbot with an ML model will determine by photo whether the product is fresh, whether it can be sold, or whether it needs a discount. During the development, VisionLabs tested the Luna Line platform to identify and classify defects in metallurgical production, search for bus stops with broken glass, and check for identification numbers on cars.

We have noticed that the market lacks ML development solutions that are accessible and understandable even without knowledge in IT. Therefore, we created Luna Line, a no-code tool with a simple interface that allows you to run ML models and train them for any industry. For example, in agriculture, with the help of Luna Line, you can assess the state of plants and animals, in industry - find defects on production lines, in education - teach schoolchildren how to work with neural networks. Our platform turns complex technologies into a service available to everyone, "said Dmitry Markov, CEO of VisionLabs.