DeepSeek-R1 · AIME(2025; Pass@1; DeepSeek-R1 paper Table 13 and accompanying prose) 75% · 附件证据(deepseek-r1-ev-009)

DeepSeek-R1 的 AIME · 2025; Pass@1; DeepSeek-R1 paper Table 13 and accompanying prose 官方出处。DeepSeek-R1 reported reasoning configuration · 75%。见《DeepSeek-R1 paper》PDF p.57; Table 13; full header, target row and explicit prose。证据 deepseek-r1-ev-009。

原文与定位

DeepSeek-R1 · AIME(2025; Pass@1; DeepSeek-R1 paper Table 13 and accompanying prose),DeepSeek-R1 reported reasoning configuration配置,Pass@1 75%(具体成绩公布时间未确认)。见《DeepSeek-R1 paper》PDF p.57; Table 13; full header, target row and explicit prose。 查看成绩条件

Average Score | AMC 12 2024 | AIME 2025 | USAMO Index | DeepSeek R1 | 143.7 | 11.3/15 | 256.7 | DeepSeek-R1 achieves a 75% solve rate (Pass@1) on AIME 2025.
官方材料
DeepSeek-R1 paper · PDF 第 57 页
定位
附件 · PDF p.57; Table 13; full header, target row and explicit prose
材料版本
Official DeepSeek-R1 paper: 86/86 PDF pages read; Table 3, Table 8, Table 13, Table 14 and Table 15 checked. Safety, jailbreak and attack sections pp.44-54 were retained as local audit only and excluded from regular mappings.
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-R1 · AIME(2025; Pass@1; DeepSeek-R1 paper Table 13 and accompanying prose),DeepSeek-R1 reported reasoning configuration配置,Pass@1 75%(具体成绩公布时间未确认)。官方出处 DeepSeek-R1 paper(附件,Official DeepSeek-R1 paper: 86/86 PDF pages read; Table 3, Table 8, Table 13, Table 14 and Table 15 checked. Safety, jailbreak and attack sections pp.44-54 were retained as local audit only and excluded from regular mappings.)PDF p.57; Table 13; full header, target row and explicit prose,https://arxiv.org/pdf/2501.12948#page=57。Benchlog https://benchlog.lekshome.cc/evidence/r1/deepseek-r1-ev-009/

关联成绩与条件

DeepSeek-R1 · AIME 成绩与配置
指标与成绩配置与评测条件证据关系
Pass@175%原文 75% · 单位 %
公布 未确认
DeepSeek-R1 reported reasoning configuration
设置
Maximum generation length 32,768 tokens. For sampled benchmarks: temperature=0.6, top_p=0.95, 64 responses per query to estimate Pass@1.
未知项
Per-benchmark sampling use is not itemized beyond the card preface.
工具
工具条件未记录
推理预算
输出上限 32768 tokens
框架
未记录
协议
DeepSeek-R1 paper Table 13 and prose: AIME 2025 solve rate is reported as Pass@1.
条件
DeepSeek R1 row and accompanying sentence explicitly report 75% solve rate.

直接支持该成绩

在排名中查看此成绩