Unverified50% confidenceBenchmarkExact time
Claude Opus在OSWorld 1.0上能达到83.5%的准确率
1
Sources
50%
Confidence
Long-term
Relevance
7/9/2026
First Seen
Sources
OSWorld 2.0测评:最强AI Agent完成率仅20%,长流程是致命短板
bilibili熊二等兵7/1/2026
Related Claims
VerifiedClaude 4.5 Opus 在 SWE-Bench Verified 上达到了 80.9% 的成绩73% similarUnverifiedClaude Opus 4.8 reached 85.9% on the GraphWalks 256K subset (up from 76.9% for 4.7) and 68.1% on the full 1-million-token version, nearly double Opus 4.7's score of 40.3%71% similarUnverifiedOpus 4.6在OS World Verified基准测试中得分72.7%70% similarUnverifiedClaude Opus 4.8在OnlineMind2Web网页操作基准测试上取得了84%的成绩68% similarUnverifiedOpus 4在SWE-bench Verified上的通过率超过72%,显著领先于其他竞争模型66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/338168API
curl https://kongchang.com/api/v1/knowledge/claims/338168MCP
get_claim(id=338168)