cHHillee argues posttraining-as-a-service is resilient: GPU rentals can't be replaced
cHHillee · x · 2026-09-22
Responding to PatrickToulme's claim that future models could rewrite the stack in an afternoon, cHHillee argues posttraining-as-a-service is quite resilient to AI improvement: renting GPUs obviously can't be replaced by a model, and building large-scale RL posttraining infrastructure requires substantial GPU-hours and token investment that can't be conjured from a single code generation.
More from Infra
- More inference than training machines means writing will soon beat reading — GregoryDiamos · 2026-09-22
- SayGm Routes Prompts Through Two Attested TEEs With NEAR AI's Confidential Compute — richdotca · 2026-09-22
- COLIBRI: pure-C zero-dep engine streams 2.8T-param MoE models from disk on consumer hardware — bibryam · 2026-09-22
- Sentdex benchmarks openjev: 169ms on Dell GB10 vs 137ms on RTX 3090 — Sentdex · 2026-09-22
- PyroDash cuts inference cost 96% by having a 4B model call the big one only when needed — jiqizhixin · 2026-09-22
- Underdog's Husky Inference Engine Claims 4.5x Speedup Over MLX, 730 tok/s on MacBook — jimmykoppel · 2026-09-22