Anthropic’s Claude Fable 5 has taken the top spot on the Epoch Capabilities Index for the first time since 2024, beating OpenAI’s GPT-5.5 Pro in the process.
Epoch AI lists Claude Fable 5 at Rank 1 out of 198 models with an Epoch Capabilities Index score of 161. GPT-5.5 Pro is listed at Rank 2 with a score of 160.
Epoch AI said Claude Fable 5 achieved a new high score on the index, beating GPT-5.5 Pro by one point. The update marks the first time Anthropic has taken the lead on the Epoch Capabilities Index in over a year.
The last major Anthropic model linked to that earlier lead was Claude 3.5 Sonnet, which launched in June 2024.
The Epoch Capabilities Index combines scores from many different AI benchmarks into one general capability scale. Epoch AI says the index is designed to compare models across longer periods, especially when individual benchmarks become saturated.
The score does not measure only one specific skill. Instead, it reflects performance across several benchmark areas.
Anthropic launched Claude Fable 5 on June 9, 2026. The company described it as a Mythos-class model made safe for general use.
Anthropic said Fable 5 is its most capable generally available model and performs strongly in software engineering, knowledge work, vision, scientific research and longer complex tasks.
The company also said some requests in sensitive areas may fall back to Claude Opus 4.8 because of added safeguards.
Epoch AI also said Claude Fable 5 performed strongly on FrontierMath. The model scored 87% on FrontierMath Tiers 1–3 and 88% on Tier 4. Epoch said these results continue Anthropic’s rapid gains in math performance.
The benchmark lead comes during a complicated launch period for Anthropic. Anthropic later said access to Claude Fable 5 and Claude Mythos 5 was unavailable after a US government directive.
The company said it was suspending access to both models and working to restore access as soon as possible.
Claude Fable 5’s new top position puts Anthropic ahead of OpenAI on Epoch AI’s general capability ranking. The margin is narrow, with Fable 5 leading GPT-5.5 Pro by one ECI point.
However, the result is still significant because it gives Anthropic the top position on a major AI benchmark after more than a year without the lead.
Get the latest tech news, telecom insights, and product launches wherever you prefer.
Add ProPakistani to Preferred Sources and see more of our stories in Google Search and Top Stories.