PALO ALTO, California — Sept. 30, 2026
Executive Summary
Topsort has introduced T-Brain, its first Large Commerce Model (LCM), at Topsort Fall Update 2026. The commerce-specific foundation model is designed to interpret signals from products, catalogs, queries, shoppers and transactions and apply that intelligence across search, retrieval, ranking, recommendations, advertising and agentic commerce. T-Brain is already in production with select existing customers, with preliminary deployments showing improvements in retrieval and conversion.
Announcement Overview
T-Brain is designed as a shared intelligence foundation for modern commerce systems, adapting to an enterprise’s catalog, shopper behavior and commerce environment. Topsort stated that the model builds on intelligence developed through billions of commerce decisions across more than 100 marketplaces, extending the company’s machine-learning work beyond sponsored-product advertising. In early production deployments, preliminary results showed a 21% improvement in retrieval and a 26% improvement in conversion on commerce surfaces where T-Brain has been applied. The model is being demonstrated at Topsort Fall Update 2026 in Boston.
Key Announcement Details
- Announcement Type: Product Launch
- Announcing Entity: Topsort
- Announcement Date: September 30, 2026
- Dateline: PALO ALTO, California
- Announcement: Topsort unveils T-Brain
- Product: T-Brain Large Commerce Model
- Category: Commerce AI
- Primary Use: Enterprise decision-making
- Applications: Search, retrieval, ranking, recommendations, advertising, agentic commerce
- Deployment: Select existing customers
- Early Results: 21% retrieval improvement; 26% conversion improvement
- Scale: Billions of commerce decisions across 100+ marketplaces
- Event: Topsort Fall Update 2026
- Location: Boston
- Leadership: Regina Ye and Francisco Larrain
- Company Focus: AI-native commerce monetization infrastructure
T-Brain Introduces a Commerce-Specific Foundation Model
At Topsort Fall Update 2026, the company introduced T-Brain, its first Large Commerce Model (LCM).
T-Brain is a commerce-specific foundation model designed to understand signals generated throughout the commerce environment, including:
- Products
- Catalogs
- Queries
- Shoppers
- Transactions
The model is designed to apply this intelligence across multiple commerce decisions rather than being limited to a single application.
Its stated use cases span:
- Search
- Retrieval
- Ranking
- Recommendations
- Advertising
- Agentic commerce experiences
Topsort stated that T-Brain is designed to adapt to each enterprise’s catalog, shopper behavior and commerce environment, providing a shared intelligence foundation for decisions across commerce systems.
Intelligence Developed Through Billions of Commerce Decisions
T-Brain builds on intelligence developed as Topsort powered billions of commerce decisions across more than 100 marketplaces.
The company stated that its initial machine-learning work focused on a central commerce decision: which sponsored product to show.
That decision required balancing:
- Shopper relevance
- Advertiser performance
- Marketplace revenue
T-Brain extends that work beyond advertising to a broader range of commerce decisions, using the intelligence developed through those earlier commerce applications as the foundation for its Large Commerce Model.
T-Brain’s Role Across Commerce Systems
The model is designed to provide a shared intelligence layer for different commerce systems and surfaces.
Rather than being positioned around a single commerce function, T-Brain supports applications spanning discovery, merchandising, advertising and emerging agentic experiences.
The company stated that the model can adapt to the specific characteristics of an enterprise, including its product catalog, shopper behavior and wider commerce environment.
This architecture is intended to allow the same commerce-specific intelligence foundation to support multiple decision-making applications.
Topsort’s Perspective on AI and Commerce
Regina Ye, CEO and Co-founder of Topsort, said the company expects AI to reshape online commerce and described T-Brain as the foundation for its approach to that change.
“AI has changed the technology landscape quickly, and commerce is entering a period of significant change as well,”
Ye added:
“I expect the online store to look fundamentally different in the next 12 to 18 months. As AI reshapes commerce category by category, businesses will need more than another assistant added to existing software. They will need intelligence that understands how their business works and helps their systems make better decisions. T-Brain is our foundation for building toward that future.”
Large Commerce Model Development
Francisco Larrain, Co-founder and CTO of Topsort, said the company has worked on T-Brain for two years.
“We’ve been working on this for two years. Large language models are built to understand and generate language; a Large Commerce Model learns the relationships among products, shoppers, intent and outcomes to help predict what a shopper is likely to do next,”
Larrain continued:
“With T-Brain, we’re bringing that commerce-specific intelligence into a shared model foundation that can support search, retrieval, ranking, advertising and agentic experiences.”
The distinction described by Larrain centers on the model’s focus on relationships among products, shoppers, intent and outcomes within commerce environments.
Early Production Results
Topsort stated that T-Brain is already being used in early production deployments with some of its customers.
The company reported preliminary results from commerce surfaces where the model has been applied:
- 21% improvement in retrieval
- 26% improvement in conversion
Topsort characterized these as preliminary results from early production deployments.
The company is also demonstrating T-Brain as part of Topsort Fall Update 2026.
Production Availability and Demonstration
T-Brain is currently in production with select existing customers, according to the company.
Topsort is demonstrating the model at Topsort Fall Update 2026 in Boston.
The event provides a setting for the company to demonstrate the commerce-specific foundation model and its applications across the commerce decision surfaces identified in the announcement.
About Topsort
Topsort is an AI-native monetization infrastructure company for modern commerce. The company helps retailers, marketplaces, delivery platforms and other commerce businesses build and operate monetization programs, ranging from auction-based advertising to AI-driven decisioning across commerce surfaces.
- Global reach: More than 40 countries
- Customers named in the announcement: Coles, JET, Avolta and Falabella
- Core focus: Monetization infrastructure and AI-driven commerce decisioning
- Commerce applications: Advertising, monetization and decision-making across commerce surfaces
Topsort stated that its infrastructure supports leading enterprises across its global customer base.
Media Contact
For additional information, visit topsort.com.
Source Attribution
Source: Company announcement
