Witbe VLMs run faster and cost less per test than general-purpose models and they can also be deployed inside the client's own infrastructure.

NEW YORK — September 3, 2026 — Witbe (Euronext Growth – FR0013143872 – ALWIT) today announced Witbe VLMs, a family of Vision Language Models built and trained by Witbe for video test automation on any device. A year ago, Witbe brought agentic testing to video services, with specialized agents that plan a test, run it on a real device and react to whatever appears on screen. Witbe VLMs work alongside the general-purpose and frontier models those agents have been leveraging.

Why Witbe trains its own Vision Language Models

General-purpose models are trained on the open web, which contains very little about set-top box on-screen keyboards, launcher grids, or the way a picture freezes on a five-year-old Smart TV. Witbe VLMs are trained on exactly those cases, drawing on the expertise Witbe has built testing and monitoring video services on any device, in any location, across any application.

Specializing a model on video test automation and monitoring does not cost generality. On applications the models had never seen during training, they perform on par with frontier models. What transfers between applications is how screens and interfaces behave.

What changes

• Lower latency, so a test suite finishes faster, and device control in the REC is more responsive.
• Lower cost per test, more predictable behavior, more control over upgrades.
• AI sovereignty. Deployed inside the client's own infrastructure, test data including recorded video of the client's service never leaves that environment. For operators facing data residency rules or content protection obligations, this is what makes AI in production possible at all.

What does not change is how a result is proven. Models interpret; deterministic logic validates. Every result stays anchored in KPIs measured on a real device, with recorded video evidence for every run. The AI ratio still sets how much AI each scenario uses, from 0% to 100%. General-purpose and frontier models remain available.

“A year ago we launched the first agentic testing capability for video services. Today, our clients run our Agentic SDK in production on thousands of Witbox units deployed in the field,” said Mathieu Planche, CEO of Witbe. “The next step was always the models themselves. Ours are trained on data we have gathered testing video services on real devices, in real conditions. This leads to a better cost proposition for our clients, and a faster pace of execution when the scenarios run.”

“The question we get asked is whether a purpose-trained model gives up generality,” said Yoann Hinard, COO of Witbe. “It does not. On applications our models had never seen in training, they perform on par with frontier models, because what carries across applications is how interfaces behave.”

Availability

Witbe VLMs are being rolled out. Access requires AI token provisioning, as with existing Agentic SDK AI capabilities. The majority of Witbe clients are already using agentic testing through Witbe Suite 41, so for them the tokens already purchased cover more testing, completed faster. Existing scenarios are unchanged. Deployment inside the client's own infrastructure is available, and requirements are provided on request.

At IBC 2026

Witbe VLMs will be demonstrated at IBC 2026, 11-14 September, RAI Amsterdam, Hall 1, Booth 1.C53. Witbe's peer-reviewed IBC paper on eight months of agentic AI in production, in which one engineer increased coverage 5x on smart TVs, is presented by Yoann Hinard in the Streaming Technology 2 session on Saturday 12 September.
Briefings and interviews: witbe.net/events/ibc-2026