Unverified60% confidenceBenchmarkExact time
Claude Sonnet 5 在 Computer Use 上得分 81.2%
2
Sources
60%
Confidence
Long-term
Relevance
7/12/2026
First Seen
Sources
Claude Sonnet 5深度实测:性能亮眼但定价策略暗藏玄机
bilibiliNixAI6/30/2026
Related Claims
UnverifiedClaude Sonnet 5在Terminal 2.1得分80.4,计算机使用得分81.2,GDP评估达1619并超越OPUS82% similarUnverifiedClaude Sonnet 4.6在OS World(智能体计算机使用)测试中从61.4%提升至72.5%76% similarVerifiedSonnet 5在Terminal Bench 2.1上取得80.4%的成绩,接近Opus 4.876% similarUnverified在SWE Bench Pro编程基准测试中,Sonnet 5取得63.2%,Opus 4.8取得69.2%70% similarUnverified在智能体计算机使用基准中,Opus 4.8 的 high 设置与 Sonnet 5 的 max 设置表现相当,但 Sonnet 5 消耗更多 token70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/490282API
curl https://kongchang.com/api/v1/knowledge/claims/490282MCP
get_claim(id=490282)