Unverified50% confidenceFactExact time
Cursor联合创始人Michael Truel表示Opus 4.8在CursorBench的每个努力级别上都超越了之前的Opus模型，工具调用更高效，步骤更少
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Claude Opus 4.8深度解析：越诚实越会应试的AI悖论
bilibili皮皮蟹勇闯天涯6/1/2026
Related Claims
UnverifiedCursor co-founder Michael Truel stated that Opus 4.8 outperforms previous Opus models at every effort level on CursorBench, with more efficient tool calls and fewer steps82% similarVerifiedCursor 2.0的自研Composer模型号称比同类模型快4倍71% similarUnverifiedCursor被曝出在高峰期自动降级模型68% similarUnverified一位开发者在 Cursor 中使用 Grok 4.5 生成 PRD 和 Issues，发现输出质量媲美 Claude Opus，响应速度提升约 5 倍67% similarUnverifiedCursor在用户流量低谷时借用生产环境的推理GPU来加速训练67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/48707API
curl https://kongchang.com/api/v1/knowledge/claims/48707MCP
get_claim(id=48707)