---
format: "aidr-story-markdown/v1"
id: "7fc3e07707c12cfa2e32d6dccf6189b2886cddff5d0962031b3df0d284822f17"
canonical_url: "https://aidr.today/7fc3e077?lang=en"
title: "Stop guessing auto scaling targets for TTS on Amazon SageMaker AI"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-22T16:30:06.000Z"
category: "Infra"
topics: ["amazon","inference","coding","framework"]
source_urls: ["https://aws.amazon.com/blogs/machine-learning/stop-guessing-auto-scaling-targets-for-tts-on-amazon-sagemaker-ai/","https://aws.amazon.com/blogs/machine-learning/orchestrating-sagemaker-ai-endpoint-auto-scaling-with-aws-cdk/"]
summary: "A post describes deploying Kokoro-82M as a custom container on Amazon SageMaker AI and deriving an autoscaling policy from two live-endpoint measurements, Q_max and T_total, instead of guessing thresholds."
---

# Stop guessing auto scaling targets for TTS on Amazon SageMaker AI

> [Open the canonical story](<https://aidr.today/7fc3e077?lang=en>)

**Published:** 2026-10-22T16:30:06.000Z
**Category:** Infra
**Topics:** amazon, inference, coding, framework

## Summary

A post describes deploying Kokoro\-82M as a custom container on Amazon SageMaker AI and deriving an autoscaling policy from two live\-endpoint measurements, Q\_max and T\_total, instead of guessing thresholds\.

## Sources

- [Story source](<https://aws.amazon.com/blogs/machine-learning/stop-guessing-auto-scaling-targets-for-tts-on-amazon-sagemaker-ai/>)
- [Story source](<https://aws.amazon.com/blogs/machine-learning/orchestrating-sagemaker-ai-endpoint-auto-scaling-with-aws-cdk/>)