Llama — Field Evidence
paperUnverified
Llama · Llama · praise
Field note
When Do Foundation Models Pay Off? A Break-Even Analysis of Pretrained Time Series Forecasters — Deploying a time series foundation model requires GPU infrastructure, engineering overhead, and carries no guarantee of improvement over XGBoost. We provide the first systematic break
collected 2026-07-07original 2026-07-06
Does this shift the US–China race?
Be the first to call it
Impact on the index
Benchmark win · minor
US +30China -7
Directional contribution — before recency decay and per-type diminishing returns. How it works →
Related Front
US vs ChinaLikely
U.S. FrontiervsChina Open-Weight
Likely — capability gap narrowing on common tasks
Letters from the Front
Be the first to file a report.