GPT-4's dominance lasted a year while today's top models barely survive seven weeks at the top
Quick Answer
OpenAI's GPT-4 dominated the Epoch Capabilities Index for nearly a year, significantly outpacing competitors, with Claude 3 Opus dethroning it in February 2024.
Quick Take
OpenAI's GPT-4 dominated the Epoch Capabilities Index for nearly a year, significantly outpacing competitors, with Claude 3 Opus dethroning it in February 2024. Since then, the lead has shifted 17 times, with models averaging only seven weeks at the top, indicating intensified competition and shorter dominance periods.
Key Points
- GPT-4 held the top ECI position for about a year, a record duration.
- Claude 3 Opus ended GPT-4's reign in February 2024.
- Since then, 17 lead changes have occurred among models.
- The average duration at the top has dropped to seven weeks.
- Competition has intensified, with smaller but faster capability jumps.
📖 Reader Mode
~1 min read
Jul 6, 2026
OpenAI's GPT-4 held the top spot on the Epoch Capabilities Index (ECI) for about a year. The ECI is a composite measure of language model performance. According to Epoch AI researcher Jaeho Lee, that's far longer than any model since. OpenAI's o1 held the second-longest lead at just over three months, less than a third of GPT-4's reign. Since Claude 3 Opus dethroned GPT-4 in February 2024, the lead has changed hands 17 times. The median stay at the top per model was about seven weeks.
The chart can be read two ways. It shows how long rival labs needed to match GPT-4, a genuine outlier at launch. But it also shows that competition has gotten much fiercer since. No model can hold a comparable lead anymore, and the capability jumps between transitions are faster but smaller compared to GPT-4 and the era that began with reasoning models like o1-preview in fall 2024.
— Originally published at the-decoder.com
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from The Decoder
See more →
An AI model programmed nonstop for 19 days on a single MirrorCode task that cost $2,600 to run
Epoch AI's MirrorCode benchmark reveals Claude Opus 4.7 as the leader with a 56% solve rate, reconstructing a 16,000-line toolkit in 14 hours. Despite this, all models tested struggle with the most complex tasks, highlighting limitations in current AI capabilities. The single task consumed $2,600 over 19 days, raising questions about cost-effectiveness in AI development.
