Artificial intelligence news.
Straight from the original announcements.

Meta says model templates cut ML engineering time by 92%

November 18, 2026, 17:59By the AI News Bites Editorial TeamSource: Engineering at Meta
Meta says model templates cut ML engineering time by 92%
Image: Engineering at Meta

Meta says its template-driven approach to machine learning reduced per-model engineering time by 92% over the last two years. The company used Standard Model Templates to coordinate improvements across its large ads ranking model ecosystem.

Key points
  • Meta says SMTs reduced per-model engineering time by 92%.
  • The approach shipped 5000+ technique-model combinations.
  • Offline model quality improved by 0.63% on average at neutral serving capacity.
  • Meta reports a 0.86% cumulative online lift covering 70% of models by count.
  • The workflow shipped 80 new ML techniques over two years.

Meta shifts from model-by-model development

Meta’s ads ranking systems process billions of daily requests across a diverse range of products and optimization goals. The company says the central engineering challenge has shifted from inventing techniques to propagating them across production models.

As the model fleet expanded, teams faced a growing mix of architectures, features, training settings, serving constraints and platform-specific logic. According to Meta, that fragmentation increased the engineering cost of shipping ideas, raised reliability costs because of additional corner cases, and made it harder to predict the effect of a change across the ecosystem.

Meta says adapting every technique separately for every model causes work to grow linearly with the number of models. It also limits automation because AutoML and optimization systems work best in a structured search space.

Standard Model Templates create shared building blocks

Meta’s response was to introduce Standard Model Templates, or SMTs. The approach organizes model design around a smaller set of reusable templates built from shared components, including architecture choices, feature logic, training settings and serving configurations.

The templates cover major ranking and retrieval patterns across different model sizes. They also package components for feature selection, feature interaction, model parallelism, training configuration and hardware-aware serving. Model-specific logic remains where it is needed, while common elements are represented and maintained as shared parts.

This changes the development process from per-model iteration to template-level iteration. Meta can evaluate a technique on a smaller group of representative models, then propagate it to model groups using the same template. The company says this reduces propagation complexity, supports stronger validation and guardrails, and makes representative-model results a more useful signal for the wider group.

Reported results and a broader optimization system

Over the last two years, Meta says the approach reduced per-model engineering time by 92%, shipped 5000+ technique-model combinations, improved offline model quality by 0.63% on average at neutral serving capacity, and delivered 0.86% cumulative online lift covering 70% of models by count.

Meta also says it shipped 80 new ML techniques through a multi-model optimization workflow. The process evaluates techniques across representative models, weighs aggregate performance against regression rates, and then validates candidates on held-out models before integrating them into a new template version.

The company says the system allows teams to transfer complete model designs between model families, rather than applying isolated components one at a time. It also supports scaling-law-based optimization workflows, allowing cost and performance trade-offs to be learned in template space and applied more broadly.

For Meta, the result is a shared foundation for improving a large model fleet. The company’s source also presents SMTs as a way to give agentic model-building systems structured, known-good design patterns, although the discussion provided here ends before describing the full implementation of that use case.

Related news

Business
BusinessNovember 19, 2026, 13:30

Sierra Whitepaper: How Personal Agents Could Reshape Enterprise Customer Relations

Sierra has published a whitepaper, "Personal agents and the enterprise", arguing that companies should prepare to serve customers both directly and through the personal…

Meta Launches Forum App for Facebook Groups in Indonesia
Products and toolsNovember 19, 2026, 07:05

Meta Launches Forum App for Facebook Groups in Indonesia

Meta has made Forum available in Indonesia, a dedicated app that brings Facebook Groups into one place for conversations, answers and community discussions. The app is…

AWS Launches SageMaker JumpStart Leaderboard for Model Comparisons
Products and toolsNovember 18, 2026, 23:31

AWS Launches SageMaker JumpStart Leaderboard for Model Comparisons

AWS has launched the Amazon SageMaker JumpStart Model Leaderboard, a SageMaker Studio view that compares open-weight models on quality, performance and estimated cost.…