Event ·Multi-source

Mistral Large 4 在 DeepSWE 基准测试中领先 GLM-5.3

Mistral Large 4 在 DeepSWE 基准测试中得分为 62%,超越 GLM-5.3。该模型在 Finch 金融任务基准测试中得分为 67%,为当前开源模型最佳表现。Le Chonk 在 Harvey 的法律智能体基准测试中得分为 15%。

Current read

Mistral Large 4 在多个专业领域基准测试中表现优异,金融和法律任务得分亮眼。

11 sources76° AI heatLast updated 2026/10/6 13:28:19

Evidence

11 sources
Primary sourceTestingCatalog
View source
Related source 1OpenRouter
View source
Related source 2kimmonismus
View source
Related source 3Mistral AI
View source
Related source 4Guillaume Lample (Mistral)
View source
Related source 5Julien Chaumond
View source
Related source 7Thomas Wolf
View source
Related source 8Simon Willison
View source
Related source 9Clement Delangue
View source
Related source 10Arthur Mensch
View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.