Fast Feed

Home

❯

Tech

❯

venturebeat

❯

Surprise upset: GPT-5.5 beats Claude Fable 5 on brutal new Agents’ Last Exam benchmark

Surprise upset: GPT-5.5 beats Claude Fable 5 on brutal new Agents’ Last Exam benchmark

Jun 10, 20261 min read

Summary

Failure

Summarization halted globally due to repeated key failures.

By carl.franzen@venturebeat.com (Carl Franzen)

Original Article


Created By Quantlight © 2026

  • GitHub