Meituan unveils MTFM, a unified recommendation foundation model
10/03/2026 — 10/03, 11:12·1 sources·1 reports
Story overview
On October 3, 2026, Meituan's technical team announced MTFM, a unified recommendation foundation model. According to the team, MTFM is built on MTGR and serves as a unified recommendation foundation model for Meituan's food delivery scenarios.
The team says MTFM is the first unified fine-ranking model to cover several major food delivery businesses, and it also shared how the model has been put into practice across those multi-business delivery scenarios. No details were disclosed about the model's parameter count, training data, evaluation benchmarks, or per-business performance numbers.
That is where matters stand for now: Meituan's technical team has made public how MTFM is positioned and where it is being applied, namely a recommendation foundation model derived from MTGR with unified fine-ranking across multiple food delivery businesses. Further specifics have yet to be released.
AI-generated from 1 reports · updated 2 hours ago
Latest turnMeituan proposes MTFM, a unified recommendation foundation model built on MTGR, which it says is the first to unify the ranking models of several major food-delivery businesses. The post covers how it was deployed across those scenarios.
Reports on this story headlines open the original
Meituan proposes MTFM, a unified recommendation foundation model built on MTGR, which it says is the first to unify the ranking models of several major food-delivery businesses. The post covers how it was deployed across those scenarios.
美团技术团队AI score 66
Other stories people are talking about
- 554NVIDIA launches 64GB DGX Spark desktop AI computer at $4,9998 sources
- 539RisingApple tightens macOS Full Disk Access over AI agent risks7 sources
- 400Meta open-sources Muse Gadgets firmware and SDK5 sources
- 234SurgeAmazon Weighs Moving $8 Billion of Nvidia Chips into a Financing Vehicle3 sources
- 227Anthropic launches Claude Frontier Academy with $100M3 sources
- 226SurgeHugging Face open-sources AstaBrief for fast report generation3 sources
How is heat calculated?About the methodHide
Heat counts how many independent sources covered a story in the last 48 hours: one source counts once no matter how many posts it published, decaying with a 24-hour half-life. What ranks first is what many people are talking about.
This page aggregates public feeds. Headlines and summaries are machine-organized and remain the property of the original authors; verify important facts at the source.
- Surge
- Discussion rising fast
- New
- First report within 6 hours
- Rising
- Still gathering discussion