The Amazon SageMaker Python SDK v3 now exposes generative AI inference recommendations in Amazon SageMaker AI directly in notebook workflows, also accessible through the SageMaker AI UI and Boto3 APIs. This release enables benchmarking an endpoint, generating data-driven deployment recommendations, and deploying the recommended configuration from a notebook. The recommendations automate inference optimization by benchmarking a live Amazon SageMaker endpoint against a synthetic or real-traffic workload, measuring throughput, time-to-first-token (TTFT), end-to-end latency, and more, then generating deployment recommendations ranked by cost-performance tradeoff and deploying the top-ranked configuration to a real-time endpoint. Previously, these capabilities required Amazon SageMaker Studio or constructing AWS SDK for Python (Boto3) API calls; now they are Python SDK operations. The new functionality is available under the `sagemaker.serve.ai_inference_recommender` package starting with version 3.17.0. Primary operations include `ModelBuilder.from_jumpstart_config(…)` to build a ModelBuilder from a JumpStart model ID and compute config, `start_benchmark(endpoint, …)` to run a load test, `mb.generate_deployment_recommendations(…)` to explore instance/framework configs and return ranked recommendations, `mb.deploy(…)` to deploy the top recommendation, and `ModelBuilder.from_recommendation_job(job_name)` to hydrate a ModelBuilder from a completed job. Prerequisites include installing the latest SDK with