Unverified80% confidenceFactTime unknown
Kimi K2.5在SWE（软件工程）测试中达到了Gemini 3 Pro的水平
1
Sources
80%
Confidence
Medium-term (~90 days)
Relevance
5/31/2026
First Seen
Valid until: 8/29/2026
Sources
DeepSeek OCR2、Kimi K2.5、微软Maia 200同日发布
bilibiliinfinite灵感港
Related Entities
Related Claims
UnverifiedKimi K2.5官网API版本的前端能力是国产最强，部分界面能与Gemini 3.0 Pro比肩78% similarUnverified前端生成测试中还对比了疑似Gemini 3的refer模型、Kimi KL和Haiku 4.5，GPT 5.1在指令遵循和视觉质感上处于领先75% similarUnverified测试者认为Kimi K2.6的Web Coding编码能力至少已达到Gemini 2.5 Pro的水平74% similarUnverified月之暗面科技发布了全新的Kimi K3大模型，据称一经推出便登顶评测榜首70% similarVerified在SWE-Bench Verified基准测试中，Gemini 2.5 Pro仅次于Claude 3.7 Sonnet68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/3913API
curl https://kongchang.com/api/v1/knowledge/claims/3913MCP
get_claim(id=3913)