Unverified50% confidenceBenchmarkExact time
LunCat 2.0是美团自研大模型旗舰版本，在C-Eval、CMMLU等中文基准上据称已接近DeepSeek V3水平
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/6/2026
First Seen
Valid until: 10/4/2026
Sources
美团CatPaw实测：免费GLM-5.2编程工具靠谱吗？
bilibili鲲鹏Talk7/2/2026
Related Claims
UnverifiedLlama 3.1 405B、DeepSeek V3等优质开源模型在许多通用任务上已与商业旗舰模型高度接近65% similarUnverifiedMeta的LLaMA 3系列最大规模为405B，DeepSeek-V3约为671B（混合专家架构）61% similarUnverifiedLuma上线了Dream Machine 2.0 Mini，主打更快的生成速度和更稳定的输出59% similarUnverifiedDeepSeek V3和R1采用MoE（混合专家）架构，使Mac Studio或DGX Spark这类设备能以可用速度运行大模型57% similarUnverifiedMimo V2.5 Pro在DesignArena的开放权重模型排行中拿下第三名，比前代Mimo V2.5高出八个名次57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/113036API
curl https://kongchang.com/api/v1/knowledge/claims/113036MCP
get_claim(id=113036)