# AWS says Deepgram gains 11-70 percent more audio per dollar on Trainium2

_Published Monday, October 19, 2026 at 12:33 PM EDT · AI, Infrastructure · Latest · Tier 2 — Notable_

![AWS says Deepgram gains 11-70 percent more audio per dollar on Trainium2 — Primary](https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/10/12/ML-21906-featured-image.png)

Deepgram's Nova-3 speech recognition model processed 11-70 percent more audio per dollar on AWS Trainium2 than on the compute options Deepgram had been running, AWS reported. The evaluation covered offline batch transcription workloads, including contact-center analytics and medical transcription, using instance prices at the time of testing.

Deepgram and AWS ported the model from standard PyTorch code, tuned its operations and built a Rust runtime that coordinates eight model replicas on one trn2.3xlarge instance. At the batch size that maximized throughput, the Rust runtime delivered a 7.5% throughput gain over the multiprocessing Python baseline. The results concern throughput-oriented batch processing rather than live voice-agent response times.

## Sources

- [Artificial Intelligence](https://aws.amazon.com/blogs/machine-learning/how-deepgram-optimized-voice-ai-at-scale-with-aws-trainium/)

---
Canonical: https://techandbusiness.org/newswire/VA4UB72VLeFwsNdJzWhWXv
Published: 2026-10-19T16:33:04.704Z
Story chronology: 2026-10-19T16:23:22.000Z
Retrieved: 2026-10-19T19:21:06.769Z
Publisher: Tech & Business (techandbusiness.org)