Unverified50% confidenceBenchmarkExact time
Opus 5在ARC-AGI 3上的得分是次优模型的三倍
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/23/2026
First Seen
Valid until: 11/21/2026
Sources
Related Entities
Related Claims
UnverifiedClaude Opus 5在ARC-AGI-3上得分30.2%，而上一代仅1.5%79% similarUnverified千问3.6在多回合智能体增强能力等维度上接近甚至部分指标高于Claude 4.5 Opus70% similarUnverifiedPrime Agent 搭配 Opus 5 模型在 ARC-AGI-3 基准上取得了 95.5% 的成绩，据称超越了报告中的人类专家基线70% similarUnverified在ARC-AGI-3上Sol Max达到接近8%，而Opus和Gemini均低于1%69% similarUnverifiedOpus 5在Frontier-Bench上的表现超过Opus 4.8的两倍69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/789614API
curl https://kongchang.com/api/v1/knowledge/claims/789614MCP
get_claim(id=789614)