Mistral AI·· 17 h ago
Introducing Timestral
Introducing Timestral
AI summary
State-of-the-art probabilistic forecasting in four sizes (5M–209M params), CPU-efficient, zero-shot ready.
Selection record
Threshold 60Official, first-handFirst 71Second 62
AdmittedSum of both 133 ≥ twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed：Mistral发布时序基础模型Timestral，属AI模型
- Why it was chosen
- Mistral brings the foundation-model approach to time series for the first time, with four sizes and CPU inference latency, showing what zero-shot forecasting costs to deploy.
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: Mistral AI · mistral.ai