# dev.to: Three Gemma 4 Deployments on One T4G for Under $3: What the Runtime Changes, and What It Doesn't — retained record

> Retained dev-ecosystems record from dev-ecosystems:devto.

Canonical: https://highsignal.app/data/records/778e0fd153ae4642

[← sources](https://highsignal.app/data)


source: [dev-ecosystems](https://highsignal.app/data/dev-ecosystems) (dev-ecosystems:devto)  published 2026-09-01 18:25 ingested 2026-09-03 02:33

[original source ↗](https://dev.to/gde/three-gemma-4-deployments-on-one-t4g-for-under-3-what-the-runtime-changes-and-what-it-doesnt-jo3)

## retained text



```
vLLM, JAX and PyTorch serving the same Gemma 4 E2B checkpoint on the same AWS G5g GPU, on one harness and one statistic. The decode ranking reverses on boot time. Nineteen instances, four and a half instance-hours, under $3 - which is what let five wrong claims get caught.

tags: aws, machinelearning, benchmarking, python

reactions=7 | comments=1
```



document fetched 2026-09-07 07:11
