{"slug":"aws-samples-sagemaker-benchmark","source_name":"aws-samples/sagemaker-benchmark","name":"AWS Samples/Sagemaker Benchmark","description":"Run a managed performance benchmark against a deployed Amazon SageMaker AI endpoint using SageMaker AI inference benchmarking (part of optimized GenAI inference recommendations; NVIDIA AIPerf under the hood). Measures TTFT, ITL, request-latency percentiles, and throughput. Use when the user asks to benchmark / load-test / measure the performance of a live endpoint.","version":1,"lift":{"pass_rate_delta_pts":47.83,"pass_rate_pct":78.3,"total_cases":23,"passed_cases":18,"tokens_delta_pct":51.7,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-04T12:04:27.935578+00:00"},"skill_score":0.7826,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":47.83,"with_pass_pct":78.3,"without_pass_pct":30.4,"tokens_delta_pct":51.7,"turns_delta_pct":0,"total_cases":23,"cases_aggregated":21,"verdict":"mixed","never_hurt":true,"completed_at":"2026-08-04T12:04:27.935578+00:00","run_id":"18a671c0-39cc-4d9e-85c3-713468d6b06c","version_number":1,"is_latest_version":true,"gate":null}],"trust":{"skill_safety":"passed","safety_status":"clean","intent_verdict":"safe","content_status":"clean","indexable":true},"license":"MIT-0","install_count":0,"manifest_hash":"ebb08479e64be6c4f23e2b083d5b0e095f8011c9d84f7da7c987fea16f66f978","raw_url":"https://app.decimal.ai/s/aws-samples-sagemaker-benchmark/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/aws-samples-sagemaker-benchmark"}