[{"data":1,"prerenderedAt":12},["ShallowReactive",2],{"stack-tr-en-vertex-ai":3},{"tagline":4,"description":5,"strengths":6,"useCase":11},"Google Cloud's managed ML platform, from custom training to Gemini models.","Vertex AI brings together notebooks (Workbench, Colab Enterprise), custom or AutoML training, Vertex AI Pipelines (managed Kubeflow Pipelines), the Model Registry, the Feature Store and online and batch prediction endpoints with drift monitoring, all in one service. Model Garden gives access to Gemini, open models (Gemma, Llama) and their fine-tuning, and everything plugs natively into BigQuery and Cloud Storage. It's our choice when the data already lives in BigQuery, or when a project combines classic ML and generative AI on Google Cloud.",[7,8,9,10],"Model Garden: Gemini, Gemma, Llama and partner models, fine-tuning included.","Pipelines: managed Kubeflow Pipelines, no cluster to run, artifact lineage.","Feature Store and BigQuery ML: features stay in the warehouse.","Online and batch endpoints with built-in drift monitoring and explainability.","ML on BigQuery data, Gemini and RAG applications, forecasting, vision and recommendation.",1792851799550]