The Definitive Guide toAI Data Centers
Ask the GuideAboutAccount
Guide › Glossary › LLMflation

LLMflation

The rapid collapse in the cost to serve a given level of AI capability as models and hardware improve.

Current numbers

~9–900x/yr by thresholdLLMflation: inference cost decline at fixed quality — ~10x/yr on a16z's original GPT-3-level basis, ~50x/yr on Epoch's Mar-2025 cross-benchmark median, ~9–900x/yr by capability thresholdas of 2024-2025 · register ↗

← All terms