知乎 on MSN
如何评价Deepseek-V4-Flash正式版?
DeepSeek-V4-Flash-0731 的模型结构、尺寸和 DeepSeek-V4-Flash-preview 是一致的,仅重新进行了后训练。跑分数据: Terminal Bench 2.1:82.7(V4-Pro-Preview 只有 72.1)NL2Repo:54.2DeepSWE:54.4Toolathlon Verified:70.3内部 ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results