DeepSeek-R1-0528 · SWE-bench Verified(Agent mode; task snapshot unspecified) 57.6% · 主文证据(ev-r1-0528-009)

DeepSeek-R1-0528 的 SWE-bench Verified · Agent mode; task snapshot unspecified 官方出处。DeepSeek-R1-0528 reported reasoning configuration · 57.6%。见《DeepSeek-R1-0528 official Hugging Face model card》Official model card / table headed “| Category | Benchmark (Metric) | DeepSeek R1 | DeepSeek R1 0528” / complete row containing “| | SWE Verified (Resolved) | 49.2 | 57.6”。证据 ev-r1-0528-009。

原文与定位

DeepSeek-R1-0528 · SWE-bench Verified(Agent mode; task snapshot unspecified),DeepSeek-R1-0528 reported reasoning configuration配置,Resolved 57.6%(具体成绩公布时间未确认)。见《DeepSeek-R1-0528 official Hugging Face model card》Official model card / table headed “| Category | Benchmark (Metric) | DeepSeek R1 | DeepSeek R1 0528” / complete row containing “| | SWE Verified (Resolved) | 49.2 | 57.6”。 查看成绩条件

| Category | Benchmark (Metric) | DeepSeek R1 | DeepSeek R1 0528 | | SWE Verified (Resolved) | 49.2 | 57.6
官方材料
DeepSeek-R1-0528 official Hugging Face model card
定位
主文 · Official model card / table headed “| Category | Benchmark (Metric) | DeepSeek R1 | DeepSeek R1 0528” / complete row containing “| | SWE Verified (Resolved) | 49.2 | 57.6”
材料版本
HF revision 4236a6af538feda4548eca9ab308586007567f52; current card snapshot; read in full
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-R1-0528 · SWE-bench Verified(Agent mode; task snapshot unspecified),DeepSeek-R1-0528 reported reasoning configuration配置,Resolved 57.6%(具体成绩公布时间未确认)。官方出处 DeepSeek-R1-0528 official Hugging Face model card(主文,HF revision 4236a6af538feda4548eca9ab308586007567f52; current card snapshot; read in full)Official model card / table headed “| Category | Benchmark (Metric) | DeepSeek R1 | DeepSeek R1 0528” / complete row containing “| | SWE Verified (Resolved) | 49.2 | 57.6”,https://huggingface.co/deepseek-ai/DeepSeek-R1-0528。Benchlog https://benchlog.lekshome.cc/evidence/r1-0528/ev-r1-0528-009/

关联成绩与条件

DeepSeek-R1-0528 · SWE-bench Verified 成绩与配置
指标与成绩配置与评测条件证据关系
Resolved57.6%原文 57.6 · 单位 %
公布 未确认
DeepSeek-R1-0528 reported reasoning configuration
设置
Maximum generation length 64K tokens; temperature=0.6, top_p=0.95; 16 responses per query for sampled Pass@1 estimates.
未知项
未另记
工具
工具条件未记录
推理预算
输出上限 64K tokens
框架
Agentless
协议
未说明
条件
Official comparison table; current-model column is DeepSeek R1 0528.

直接支持该成绩

在排名中查看此成绩