Unverified80% confidenceFactTime unknown
在GPQA Diamond（博士级推理）测评中，Claude Opus 4.7以94.2%夺冠，DeepSeek V4 Pro以90.1%拿下开源第一
1
Sources
80%
Confidence
Medium-term (~90 days)
Relevance
5/31/2026
First Seen
Valid until: 8/29/2026
Sources
DeepSeek V4 Pro深度评测：对比8款旗舰模型谁更值得用
bilibili磊哥哥哈哈哈哈
Related Entities
Related Claims
UnverifiedClaude 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%74% similarUnverified在Diamond子集中，Claude 3.0 Opus得分5.2%，Gemini 1.5 Pro得分4.7%，GPT-4o Mini得分4.6%，Kimi K2.6得分3.8%73% similarUnverifiedClaude Opus 4.8在智能体实战跑分中拿下1890分，比上一代高出137分，甩开GPT 5.5达121分72% similarPartially VerifiedClaude Opus 4.8在SWE-bench基准测试中得分69.2%，GPT 5.5为58.6%，Google Gemini为54.2%71% 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69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/10498API
curl https://kongchang.com/api/v1/knowledge/claims/10498MCP
get_claim(id=10498)