# NVIDIA open-sources Kumo Forecast time-series model under Apache 2.0

> **Key takeaway:** NVIDIA released Kumo Forecast, a 340-million-parameter multivariate time-series forecasting model with an Apache 2.0 license, available through Hugging Face with pretrained weights and a DataFrame-first Python API.

**Published:** 2026-10-21T12:51:24+00:00 
**Source:** Hugging Face Blog & Announcements 
**Category:** ai-open-models 
**Canonical URL:** https://fosswire.org/news/nvidia-open-sources-kumo-forecast-time-series-model-under-ap.html 

## Summary
NVIDIA Kumo Forecast (formerly NV-Tesseract Forecasting) is now available as open source under Apache 2.0. The model combines a frozen MOMENT-1-large Transformer encoder with a trainable forecasting head, and offers an optional retrieval-augmented mode called DARR that blends direct predictions with historical pattern matching. It ships with interpretability tools and a pandas-native API, with model weights and code hosted on Hugging Face.

## Architecture and retrieval mode
- **Backbone**: MOMENT-1-large Transformer encoder (~340M parameters), kept frozen during fine-tuning
- **Forecasting head**: Trainable layer maps representations to future values; supports autoregressive extension for longer horizons
- **DARR mode**: k-nearest-neighbor retrieval from a separate historical context DataFrame, blended with direct forecast via configurable weights
- **Input constraints**: Numeric temporal data only; sequence lengths of 256, 512, 1024, or 2048 observations
- **Interpretability**: Horizon-specific lag/feature attributions, semantic-flow and trajectory-stability diagnostics, optional embedding Integrated Gradients
- **Distribution**: Model code at `NVIDIA/Kumo-TS`, weights at `nvidia/Kumo-Forecast` on Hugging Face

## Why it matters
The release adds a commercially permissive, pretrained time-series foundation model to the open AI ecosystem, with retrieval augmentation and built-in interpretability features that are uncommon in existing open forecasting tools.