TabPFN-2.5
Pre-trained tabular foundation model that turns tables into predictions for classification, regression, forecasting and anomaly detection without tuning or ML pipelines
The product is a pre-trained tabular foundation model for predictions on structured data that delivers classification, regression, forecasting and anomaly detection without tuning or ML pipelines. It sells to developers and data teams in B2B organizations across tech, finance, healthcare, energy and industrials. It is delivered as SaaS and infrastructure API with deployment via Databricks, AWS SageMaker and Microsoft Azure and is available with free tier and enterprise options.
Key features
- Classification for structured data
- Regression for structured data
- Time-series forecasting
- Anomaly detection
- Supervised and unsupervised prediction
- Deploy via Databricks Lakehouse
- Deploy via AWS SageMaker
- Deploy via Microsoft Azure AI Foundry
- Model Context Protocol for agents
- LinkedIn0.6/day
- X0.6/day
- Reddit0.1/day
GTM channels
- Blog
- Marketplace
- Community
- API
- Docs
- Changelog
ICP
- Data analytics teams
- Software developers
- Engineering teams