Unverified50% confidenceBenchmarkExact time
Previously, top models such as GPT-4 had scores on ARC-AGI that hovered in the range of single digits to 30 for a long time.
1
Sources
50%
Confidence
Long-term
Relevance
9/21/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOn the ARC-AGI-3 benchmark, the same GPT-5.6 model with only two configuration items in the Codex Harness changed saw its score rise from 13.3% to 38.3%, with output tokens reduced by a factor of six.71% similarUnverifiedThe best-performing Edu-QuRater achieved an average accuracy of 0.917 in reproducing GPT-4.1-mini pairwise judgments on the held-out set.70% similarUnverified在GPT-4发布初期，主流模型在GPQA上的准确率约为35-40%，人类领域专家基准线约为65%，随机猜测为25%70% similarUnverifiedIn EvoArena testing, GPT-5.5, which achieved the highest Step Level score, had an accuracy of just over 50%, while Gemini-3.1 Pro and GLM-5.1, which performed best at the Chain Level, reached only approximately 28%–29%.68% similarUnverifiedOpus 5在ARC-AGI 3上的得分是次优模型的三倍68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/938885API
curl https://kongchang.com/api/v1/knowledge/claims/938885MCP
get_claim(id=938885)