Mistral 发布 Timestral 时序预测模型，209M 参数超越 Google TimesFM-3
Mistral AI Releases Timestral: A 209M Forecasting Model Family That Outperforms Google’s TimesFM-3
Mistral 做了个时序预测模型 Timestral，CPU 就能跑，zero-shot 成绩还超过了 Google 的 TimesFM-3，做预测的可以试试。
Mistral 发布时序预测基础模型 Timestral，参数规模从 5M 到 209M 共多个档位。209M 版本在 GIFT-Eval、TIME 和 fev-bench 三个基准上的 zero-shot 成绩领先，超过 Google 的 TimesFM-3。该模型可以跑在 CPU 上，不需要 GPU，部署成本低。适合销售预测、库存规划等传统时序场景直接拿来用。
Mistral AI Releases Timestral: A 209M Forecasting Model Family That Outperforms Google’s TimesFM-3
Mistral released Timestral, a time series foundation model family from 5M to 209M parameters. The 209M model leads GIFT-Eval, TIME and fev-bench among zero-shot models and runs on CPU. The post Mistral AI Releases Timestral: A 209M Forecasting Model Family That Outperforms Google’s TimesFM-3 appeared first on MarkTechPost .
