AI cost to match fixed benchmark falls 47% per quarter, 725x cheaper in 18 months: Epoch
rohanpaul_ai · x · 2026-09-24
Epoch AI research shows AI is getting cheaper faster than any transformative tech in history: since 2023, the cost to reach a given benchmark performance has fallen 47% per quarter, roughly 13x per year across five primary benchmarks.
- Case in point: On GPQA Diamond, OpenAI's o3 needed about $0.30 per question for a 75% score; under 18 months later, GPT-5.6 Luna matched that level for $0.0004 — a 725x cost drop.
- Cross-tech comparison: The decline is 4x faster than DNA sequencing, 6x faster than compute, 18x faster than lithium batteries, and 54x faster than electricity (up to 1973).
- Methodology: Unlike token-price studies, Epoch measures dollars needed to hit the same benchmark accuracy, estimating each model's cost-performance curve by replaying benchmark transcripts under tighter token budgets, using a method adapted from CAIS.
Related event: Epoch AI: Cost of Equivalent AI Performance Falls 47% Per Quarter(13 posts)→
More from Infra
- Open-source ai-memory gives Claude Code, Codex, Cursor and 20+ AI coding tools a shared memory — alex_verem · 2026-10-19
- OpenRouter Data Shows US Model Token Use Flat for 8 Weeks as Chinese Providers Hit 79.2T Weekly — SumitGup · 2026-10-19
- CMU seminar digs into Floe lakehouse engine: no buffer pool, I/O faster than io_uring — sh_reya · 2026-10-19
- Data centers use just 0.85% of global primary energy, by the numbers — DarinFeinstein · 2026-10-19
- vLLM 0.32.0 adds structured-decision API reading first-token probabilities — AccBalanced · 2026-10-19
- GPU pricing is now set by cost of capital, not spot rates, argues Maya — AccBalanced · 2026-10-19