What changed: Mistral AI released Timestral, its first family of time-series foundation models. The models forecast related series together and express uncertainty in their predictions. A published GitHub repository lets organisations run them on their own data.
Who is affected
Businesses using time-series forecasts for capacity, supply chains and maintenance, along with organisations adding forecasting to existing monitoring pipelines.
What it means for your business
For businesses using Timestral, forecasts can incorporate related sensor readings and known future inputs such as production schedules. Mistral Forge also supports fine-tuning on an organisation’s own data.
What to do
Run Timestral on your own time-series data using the published GitHub repository; compare its forecasts with the results used in your existing planning process.
When it takes effect
Announced on October 15, 2026.
