Artificial Analysis· @ArtificialAnlys · X·· 2 小时前AI 评分44
AI 导读
在 Mistral Vibe 中运行时,Mistral Large 4(high)在 Artificial Analysis Coding Agent Index 上以 48 分位列当前开源权重模型第二,介于 Opencode 中的 GLM-5.3(max)(49 分)与 DeepSeek Harness 中的 DeepSeek V4.1 Flash 1017(max)(46 分)之间,而每任务总 token 用量约为后者的一半(8.5M 对 15.9M 和 16.7M)
正文
Running in Mistral Vibe, Mistral Large 4 (high) ranks second of the current open weights models on the Artificial Analysis Coding Agent Index (48), between GLM-5.3 (max) in Opencode (49) and DeepSeek V4.1 Flash 1017 (max) in DeepSeek Harness (46), while using about half as many total tokens per task (8.5M vs. 15.9M and 16.7M)
来源:Artificial Analysis · x.com