# SalesforceAIResearch/uni2ts

Unified Training of Universal Time Series Forecasting Transformers

Repository: https://github.com/SalesforceAIResearch/uni2ts
Canonical: https://ross.abutalabs.com/products/uni2ts
Language: Jupyter Notebook
License: Apache-2.0
License Family: permissive
Topics: deep-learning, forecasting, machine-learning, pre-trained-models, pre-training, representation-learning, time-series, time-series-forecasting, transformers, universal-forecasting
Last push: 2026-06-02T18:53:33+00:00

## Health v2 (maintenance only)
Score: 60/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 85, release rhythm 23, longevity 67
- inputs: {"age_days": 938, "days_push": 92, "days_rel": 302, "gap_med": 340, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1580, forks 214 (observed 2026-08-28T04:05:06.876958+00:00)

## What it is
Uni2TS is a PyTorch library for unified pre-training, fine-tuning, inference, and evaluation of universal time series forecasting transformers like Moirai and Moirai-MoE. It provides a research framework with pre-trained model weights and benchmarking utilities.

## Use cases
- forecast future values of time series with pre-trained transformer models
- fine-tune a universal time series foundation model on my own dataset
- run zero-shot time series forecasting without training a model
- evaluate forecasting models on Monash, LSF, and PF benchmarks
- pre-train a time series transformer at scale
- compare Moirai, TimesFM, Chronos, and VisionTS forecasters

## When to choose
- you need state-of-the-art zero-shot or fine-tuned time series forecasting
- you are researching universal time series transformer architectures
- you want pre-trained Moirai model weights with PyTorch tooling

## When to avoid
- you need simple classical forecasting methods like ARIMA or Prophet
- your project is not Python/PyTorch based
- you need production low-latency serving rather than research workflows

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, llm-training
- domain: time-series, machine-learning, deep-learning, data-science
- platform: python
- tags: time-series-forecasting, transformers, pre-trained-models, pre-training, fine-tuning, moirai, pytorch

## Member repositories
- SalesforceAIResearch/uni2ts (main) score 60

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:06.876958+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:56:36.132904+00:00, confidence not recorded.
  - readme: https://github.com/SalesforceAIResearch/uni2ts (fetched 2026-08-28T04:05:06.876958+00:00, sha f29cfad92bff)
- Data as of 2026-08-30T08:39:29.467469+00:00.