Verified65% confidenceOpinionTime unknown
M2.5在编程任务上跑出了接近Claude Opus 4.6和GPT-5.2等旗舰级模型的表现
3
Sources
65%
Confidence
Medium-term (~90 days)
Relevance
5/31/2026
First Seen
Valid until: 8/29/2026
Sources
MiniMax M2.5实测：10B参数如何跑出旗舰级编程能力
bilibili程序员晓刘
Related Entities
Related Claims
UnverifiedSWE-1.7被宣称在编程智能水平上已接近GPT-5.5和Claude Opus的表现82% similarUnverified测试的两个项目此前已用Claude Opus 4.5加GPT-5.5的组合写过一轮80% similarVerifiedGPT-5.5在多项能力上全面超越Claude 4.7和Gemini 3.1 Pro等模型77% similarUnverifiedGPT 5.5被用于自动配置Claude Opus 4模型，实现'用AI配置AI'77% similarPartially Verified本次测评对Claude Opus 4.8、GPT 5.5、MiniMax M3、Mimo 2.5 Pro和DeepSeek V4 Pro五个模型进行了横向比较77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/13430API
curl https://kongchang.com/api/v1/knowledge/claims/13430MCP
get_claim(id=13430)