# Exowatt uses SageMaker serverless fine-tuning to beat frontier models on alarm triage and energy analysis

> Original title: Exowatt beats frontier models on accuracy with SageMaker serverless model customization

- Source: AWS Machine Learning Blog
- Published: 2026-10-27T16:50:55.000Z
- Bad Influence: https://badinfluence.news/items/l5cqiypfpesirq5t2hn2fhugc
- Original: https://aws.amazon.com/blogs/machine-learning/exowatt-beats-frontier-models-on-accuracy-with-sagemaker-serverless-model-customization/

## Summary

Exowatt used Amazon SageMaker AI serverless model customization to fine-tune Qwen3-1.7B for edge alarm triage and Qwen3-14B for cloud energy-market analysis. The fine-tuned Qwen3-1.7B scored 0.911 macro-F1 on alarm triage, above Claude Sonnet 4.6's 0.698, and the fine-tuned Qwen3-14B scored 0.975 on energy-market analysis, above Claude Opus 4.7's 0.625. The Qwen3-14B cost $0.60 per 1,000 answers versus $29.26 for Claude Opus 4.7, and the post describes deploying to edge CPUs and Amazon Bedrock Custom Model Import.

## Why it matters

The post walks through a full fine-tune, evaluate, and deploy pipeline with measured accuracy and cost against frontier models, useful for teams weighing small task-specific models at the edge.