Unverified50% confidenceBenchmarkExact time
On the GDPval-AA benchmark for human preference evaluation, Claude Opus 4.6 scored 1606 Elo and Gemini 3.1 Pro scored 1317 Elo.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/16/2026
First Seen
Valid until: 1/14/2027
Sources
Related Entities
Related Claims
UnverifiedClaude Opus 5 scored 1861 on the knowledge work benchmark GDPval, above the human average.63% similarUnverifiedGemini 3 Pro scored 31.1% on ARC-AGI-2, Claude Opus 4.6 scored 68.8%, and GPT-5.2 scored 52.9%.63% similarUnverifiedOn GDP Val 2.1, Opus 5.5 achieved an ELO score of 1846, Fable 5.1 scored 1735, and Astra scored 1542.63% similarUnverifiedClaude Opus 5.5, released by Anthropic, scored 167 on Epoch AI's ECI benchmark, ranking first in the world.61% similarUnverifiedIn the legal work test, Gemini 4 Argon scored 19.6 and Claude Opus 5.5 scored 3.8, making Argon more than five times Opus.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1046047API
curl https://kongchang.com/api/v1/knowledge/claims/1046047MCP
get_claim(id=1046047)