AI Model Reports
AI Model Industry Trends Report
Model scale, training cost, context length, enterprise spend, and current model-market composition.
Training compute growth
Epoch AI4.5x/yr
growth rate
Frontier training cost growth
Epoch AI3.5x/yr
growth rate
Context window growth
Epoch AI30x/yr
growth rate
Enterprise spend growth
Menlo Ventures3.2x YoY
growth
Key Findings
The market is shifting on scale, cost, and buyer behavior
Compute intensity, training cost, context length, and enterprise spend are moving together. Model choice should be read against those industry pressures, not only against a single benchmark score.
01
Training compute for notable AI models has grown at an estimated 4.5 times per year since 2010, per Epoch AI.
02
The estimated training cost of frontier language models has risen roughly 3.5 times per year since 2020.
03
Context windows for leading language models have expanded an estimated 30 times per year since 2023.
04
These trajectories are long-run estimates built from publicly disclosed model characteristics, not audited or standardized figures.
Source: Epoch AI Trends. All figures are estimates derived from publicly available model disclosures and are subject to revision as new data emerges.
Industry trend indicators
Externally sourced signals for scale, context length, enterprise spending, usage-study size, and organizational adoption.
