Anthropic has released Claude Opus 5, which achieved a verified 30.16% score on the ARC-AGI-3 benchmark, a significant four-fold increase over the previous best of 7.78%. This benchmark tests an AI's ability to adapt in unfamiliar interactive environments without explicit instructions. The new model features a one-million-token context window and maintains the same API pricing as its predecessor, Opus 4.8, though its default reasoning capabilities may increase operational costs for some users. AI
IMPACT Sets a new SOTA on adaptation benchmarks, potentially influencing future model development and evaluation strategies.
RANK_REASON Frontier-lab model release with system card and benchmark results. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
Read on dev.to — Anthropic tag →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
