Forecasting

Timestral

A family of models for forecasting time series with no task-specific training: from the history of sales, demand or load it gives a forecast with intervals (21 quantiles). It takes up to 32 thousand points of history and forecasts up to 1024 steps at a time. By Mistral's measurements, the best zero-shot model on GIFT-Eval, TIME and fev-bench. Runs on a CPU.

Developer
Mistral AI, France
First release
Oct 2026
Latest release
Oct 2026
Sizes
5M, 28M, 123M and 209M
License
Commercial use with conditionsModified MIT: free, but if the company's revenue for the previous month exceeds 20 million dollars, the rights under the license do not apply and an agreement with Mistral is needed
Running
On your own serverAlso runs without a GPU
Industries
Retail and marketplaces, Finance, Manufacturing and logistics, Customer support
Deploy on your server Hugging Face GitHub

What it does

  • forecasting sales and demand from history
  • forecasting load on a warehouse, a till, a call center
  • assessing the risk of stock shortage or surplus from forecast intervals
  • forecasting with factors known in advance (promotions, holidays)

Where it is used

retail and warehousesfinance and planningmanufacturingcustomer support and call centers

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
fits
1 GPUOne GPU with 16–80 GB — mid-size versions
no versions
ClusterServer with several GPUs — flagship versions
no versions

Versions

  1. Timestral-209M
  2. Timestral-123M
  3. Timestral-28M
  4. Timestral-5M

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.Hugging Face
How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.

Frequently asked questions

Can Timestral be used in a commercial project?

With conditions. License: Modified MIT: free, but if the company's revenue for the previous month exceeds 20 million dollars, the rights under the license do not apply and an agreement with Mistral is needed. Restrictions vary — region, company revenue, attribution requirements. Have a lawyer check the terms before a commercial launch.

What hardware does Timestral need?

At minimum: Laptop or regular PC, up to 8 GB of VRAM — smaller versions. Some versions also run on an ordinary CPU, without a GPU. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does Timestral support Russian?

Language does not matter for this model: it does not work with text.

Where can I download Timestral and what does it cost?

The Timestral weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.

How I deploy it for clients

  1. SelectionI pick the model size for your task and hardware and test it on your examples.
  2. DeploymentI deploy it on your server or in a closed network and provide an API.
  3. Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
  4. IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Similar models

ForecastingTimesFMGoogle · USACommercial use with conditions

A ready-made Google forecasting model: forecasts any time series without training on your data.

DetailsForecastingChronosAmazon · USACommercial use allowed

Amazon forecasting models, among the most downloaded. Chronos-2 takes external factors into account: prices, promotions, weather.

DetailsForecastingTotoDatadog · USACommercial use allowed

A Datadog forecasting model trained on server and application metrics. Especially strong for IT monitoring: load, latency, errors.

Details

Source: huggingface.co/mistralai/Timestral-209M. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.