2026-11-01

The Latest in Forecast Benchmark Hill Climbing

Researchers are harvesting the fruits of what must have been a busy summer. Google released TimesFM-3 back on August 31:

We evaluated TimesFM-3 on three comprehensive public forecasting benchmarks: Gift-Eval, FEV-Bench, and Time. On all three benchmarks, TimesFM-3 is the top-ranked model in terms of both point and probabilistic forecasting metrics among all pre-trained foundation models.

The Forecasting Company released t0-gamma on October 13:

t0-gamma is our new forecasting foundation model. It has a similar architecture to t0-beta, which our technical report describes in detail. It strengthens the t0 family’s ability to forecast from target history, related series and inputs known ahead of time. Weeks after the release of t0-beta, t0-gamma achieves state-of-the-art performance on public benchmarks. At the time of release, it ranks first among time-series foundation models on GIFT-Eval, fev-bench and TIME.

Only to be surpassed by Mistral on October 15, getting into the business of time series foundation models with Timestral:

We evaluated Timestral on the three main public forecasting benchmarks: GIFT-Eval, TIME and fev-bench. Among pre-trained, zero-shot models, it ranks first across all five metrics.

Records be tumbling.