README
the first lines# Timestral
Blog post ·
Hugging Face collection ·
License
Timestral is a family of time series foundation models for multivariate probabilistic forecasting, with support for past-only and known-future covariates.
This repository provides standalone inference for Timestral models. It needs only PyTorch, NumPy, `safetensors` and `huggingface_hub`, and runs on CPU or GPU.
> License: the code in this repository is licensed under Apache 2.0. The Timestral model weights are released separately on the Hugging Face Hub under a Modified MIT license. See License.
## Highlights
At launch, Timestral-209M ranked first among pretrained and zero-shot models on the three main public forecasting benchmarks:
- 🥇 GIFT-Eval: #1 among pretrained and zero-shot models.
- 🥇 TIME: #1 across 98 tasks.
- 🥇 fev-bench: #1 across 100 tasks.
Ask a scout
a deep read
DeepSeekDeepSeek . DeepSeek V4 Pro for the deep read
mistralai is not in a camp, so the read goes to DeepSeek, the open source scout.
10 of 10 public reads left today. Holders get 3 a day.
No read kept yet.A read is kept for a few hours, so the next visitor sees it without a new call.
Scout calls and heat are information, not advice. They describe public GitHub activity and are never a signal to trade.
Recent commits
default branch- 14dTimestral release . Artjom Joosen
Releases
newest firstNo releases.This repo has not published a release on GitHub.