whereismyai
FoundationForecast
FoundationForecast is a python API that unifies multiple time series foundation models under one common class and interface.
FoundationForecast is a python API designed to unify time series foundation models into a single interface. It provides a consistent data format and core methods like forecast, cross-validation, and anomaly detection across various models. It removes the friction of managing different labs, dependencies, and data conventions. The library supports models such as Chronos, FlowState, Moirai, PatchTST-FM, Sundial, T0, Tabby, TabPFN, Tafsut, TiRex, TimeH2O, TimeGPT, TimesFM, Timestral, and Toto. It also runs automatically on GPUs when available.
FoundationForecast at a glance
Key features
- Forecast across time series foundation models
- Run cross-validation with a unified interface
- Detect anomalies across different models
- Pass known-future exogenous variables with Xdf
- Replicate GIFT-Eval and fev-bench in CI
- Run automatically on GPU when available
Use it for
- Forecast time series data using foundation models
- Run cross-validation on time series datasets
- Detect anomalies in time series data
- Evaluate and benchmark models with GIFT-Eval
Alternatives to FoundationForecast
- Chronos — Provides state-of-the-art zero-shot performance on time series benchmarks.
- Nixtla TimeGPT — Foundational model for forecasting and anomaly detection.
- MiroFlow — Achieves top rankings on multiple agentic benchmarks with modular architecture.
- FinGPT — Access open-source financial language models and APIs at a low cost.
- Pagesift — Extracts web data with high memory efficiency and speed for AI agents.
- MiroFish — Simulates thousands of autonomous agents to forecast complex real world outcomes.
- Helix Foundry — Connects company data into a local ontology with local-first AI and no cloud accounts.
- Finetuning With Sampling — Official implementation of the Finetuning with Sampling research paper from Harvard.
- FinCatch Studio — Simulates forward scenarios and stress-tests financial models on primary facts.
- FAROS — Connects literature evidence, executable plans, and human review into an auditable research loop.
- Lieflat Charts — Provides automated, high-quality data visualization tailored for AI agents.
- Vizro — Combines low-code configuration with high-code customization for data dashboards.
All FoundationForecast alternatives