# Mature models migrate to distributed gaming GPUs

- **Speaker:** Greg Osuri
- **Attribution:** attributed
- **Said:** 2023-09-06
- **Theme:** Local Compute — https://www.gregosuri.com/predictions/local-compute/
- **Source:** "\#3 - Akash Network with Greg Osuri" (Weapons of Mass Adoption)
- **Watch at:** 00:41:37 — https://podcasts.apple.com/us/podcast/3-akash-network-with-greg-osuri/id1705930764?i=1000626951051
- **Canonical:** https://www.gregosuri.com/predictions/local-compute/2023-09-06-mature-models-migrate-to-distributed-gaming-gpus-y7l5/
- **Cite as:** Greg Osuri, "Mature models migrate to distributed gaming GPUs," "\#3 - Akash Network with Greg Osuri" (Weapons of Mass Adoption), 6 Sept 2023, 00:41:37. https://podcasts.apple.com/us/podcast/3-akash-network-with-greg-osuri/id1705930764?i=1000626951051

## Quote — verbatim

> Most advanced models like GPT-4... work well on high-end GPUs. But as they mature, as they get older, they get optimized to work on a distributed set of GPUs like 3090s, 4090s, and whatnot, the lower-end GPUs, right? Like gaming machines.

## Context — site annotation, not spoken

Answering whether "GPU stardust" — individuals' single GPUs — can earn on Akash; his thesis that model maturation pushes inference/fine-tuning down to consumer hardware over time.