Expired90% confidenceFactTime unknown
In the real-world test, both sides used Claude terminal with DeepSeek v4 and executed the same prompts.
1
Sources
90%
Confidence
Medium-term (~90 days)
Relevance
5/31/2026
First Seen
Valid until: 8/29/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedClaude Code, after using the Ultra Think prompting technique, ranked first under Claude 4.0 with a score of 717059% similarUnverifiedClaude Opus 4.8在3D射击游戏测试中获得70分评价，地图渲染偏雾蒙蒙且缺少下蹲等基础FPS操作58% similarUnverifiedTrae and Windsurf ranked at the bottom under Claude 3.7, yet entered the top three under Claude 4.0, suggesting they may have been specifically optimized for Claude 4.058% similarUnverifiedClaude Opus 4.6 leads in performance on real-world expert-level tasks in the SWE-bench benchmark.57% similarUnverifiedAhmad Awais made DeepSeek v4 outperform Claude Opus 4.7 by applying deterministic Repair Logic.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/6389API
curl https://kongchang.com/api/v1/knowledge/claims/6389MCP
get_claim(id=6389)