Announcing Artificial Analysis Intelligence Index v4.2

▤ Summary

Artificial Analysis released Intelligence Index v4.2 with new benchmarks, private test sets, and grading upgrades. The update adds AA-Briefcase for agentic knowledge work and GDP.pdf for long-context professional document reasoning. It also increases the use of held-out test data to reduce gaming and improve the realism of the benchmark. The article highlights major model rankings, including leading positions for Anthropic and OpenAI. It also previews ongoing work on Index v5 and more incremental releases ahead.

▦ Linked Companies