DeepSeek-V3.1-Terminus · SimpleQA(Internal search agent protocol; task version unspecified) 96.8% · 主文证据(ev-v3-1-terminus-009)

DeepSeek-V3.1-Terminus 的 SimpleQA · Internal search agent protocol; task version unspecified 官方出处。DeepSeek-V3.1-Terminus reported comparison · 96.8%。见《DeepSeek-V3.1-Terminus official Hugging Face model card》Official model card / table headed “| Benchmark | DeepSeek-V3.1 | DeepSeek-V3.1-Terminus |” / complete row containing “| SimpleQA | 93.4 | 96.8 |”。证据 ev-v3-1-terminus-009。

原文与定位

DeepSeek-V3.1-Terminus · SimpleQA(Internal search agent protocol; task version unspecified),DeepSeek-V3.1-Terminus reported comparison配置,Accuracy 96.8%(具体成绩公布时间未确认)。见《DeepSeek-V3.1-Terminus official Hugging Face model card》Official model card / table headed “| Benchmark | DeepSeek-V3.1 | DeepSeek-V3.1-Terminus |” / complete row containing “| SimpleQA | 93.4 | 96.8 |”。 查看成绩条件

| Benchmark | DeepSeek-V3.1 | DeepSeek-V3.1-Terminus | | SimpleQA | 93.4 | 96.8 |
官方材料
DeepSeek-V3.1-Terminus official Hugging Face model card
定位
主文 · Official model card / table headed “| Benchmark | DeepSeek-V3.1 | DeepSeek-V3.1-Terminus |” / complete row containing “| SimpleQA | 93.4 | 96.8 |”
材料版本
HF revision 19510d6dc61f79dbd925bd51ee8a9081c509a4b6; current card snapshot
版本公布
未确认
披露时间
未确认 · 首发/后补未确认
采集时间
历史依据
暂无可靠历史版本依据
引用文本

DeepSeek-V3.1-Terminus · SimpleQA(Internal search agent protocol; task version unspecified),DeepSeek-V3.1-Terminus reported comparison配置,Accuracy 96.8%(具体成绩公布时间未确认)。官方出处 DeepSeek-V3.1-Terminus official Hugging Face model card(主文,HF revision 19510d6dc61f79dbd925bd51ee8a9081c509a4b6; current card snapshot)Official model card / table headed “| Benchmark | DeepSeek-V3.1 | DeepSeek-V3.1-Terminus |” / complete row containing “| SimpleQA | 93.4 | 96.8 |”,https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus。Benchlog https://benchlog.lekshome.cc/evidence/v3-1-terminus/ev-v3-1-terminus-009/

关联成绩与条件

DeepSeek-V3.1-Terminus · SimpleQA 成绩与配置
指标与成绩配置与评测条件证据关系
Accuracy96.8%原文 96.8 · 单位 %
公布 未确认
DeepSeek-V3.1-Terminus reported comparison
设置
The table does not specify an effort setting; search-agent template/tool-set changed as stated in the card.
未知项
Do not infer a reasoning mode from the predecessor comparison.
工具
允许工具
推理预算
未记录
框架
未记录
协议
未说明
条件
Located in the Agentic Tool Use section; exact metric is not stated here.

直接支持该成绩

在排名中查看此成绩