DeepSeek-R1-Distill-Qwen-7B · 官方评测披露
DeepSeek-R1-Distill-Qwen-7B 当前收录 6 项测试的官方记录。已完成列明材料的核查。
模型发布 · DeepSeek R1 Distill Qwen
官方材料与核查范围
All listed official model-card, official News/API, metadata and directly linked technical-report materials for this release were read through the requested scope. Tables without an explicit unit retain an empty unit; figures without exact labels were not estimated. 核对时间 。
当前资料不足以还原首发。 No reliable historical snapshot proving the complete launch-day material set was obtained. Official publication dates and current materials do not substitute for a launch snapshot.
- DeepSeek-R1-Distill-Qwen-7B official Hugging Face model card · 主文 · HF revision 916b56a44061fd5cd7d6a8fb632557ed4f724f60; current card snapshot; full Markdown reviewed公布 未确认 · 采集 · 已核查 Official model-card Markdown read in full; benchmark rows, notes, configuration statements and linked reports checked.
- DeepSeek-R1-Distill-Qwen-7B official Hugging Face model card metadata · 主文 · author=deepseek-ai; repo createdAt=2025-01-20T09:16:14.000Z (repository upload time only); lastModified=2025-02-24T03:32:20.000Z; sha=916b56a44061fd5cd7d6a8fb632557ed4f724f60公布 未确认 · 采集 · 已核查 Official model API metadata checked for author, repository identity, revision and timestamps; createdAt was never used as release date.
- DeepSeek-R1 发布,性能对标 OpenAI o1 正式版 · 主文 · Official dated DeepSeek announcement; release date 2025-01-20; the announcement explicitly names this checkpoint.公布 · 采集 · 已核查 The official dated entry or current service table was read for this release scope; aliases were not treated as weight checkpoints.
- 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.公布 未确认 · 采集 · 已核查 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.
配置
- DeepSeek-R1-Distill-Qwen-7B · Distilled dense checkpoint; source card preface states max generation length 32,768 tokens, temperature=0.6, top_p=0.95, and 64 responses per query for sampled pass@1 estimates. 输出上限 32768 tokens未知项 The source does not specify a separate effort value for each distilled checkpoint.
直接前代与后续发布
当前收录未确认直接前代。 官方关系依据
测试记录 · 6
- AIME · 2024 edition; Pass@1; table does not print a percent sign本次发布有 1 条成绩记录,含 1 条精确值(可含历史记录)
- AIME · 2024 edition; Cons@64; table does not print a percent sign本次发布有 1 条成绩记录,含 1 条精确值(可含历史记录)
- MATH-500 · 500-question set; Pass@1; table does not print a percent sign本次发布有 1 条成绩记录,含 1 条精确值(可含历史记录)
- GPQA Diamond · Diamond subset; task snapshot unspecified; table does not print a percent sign本次发布有 1 条成绩记录,含 1 条精确值(可含历史记录)
- LiveCodeBench · Version unspecified; Pass@1; table does not print a percent sign本次发布有 1 条成绩记录,含 1 条精确值(可含历史记录)
- Codeforces · Rating; contest snapshot unspecified; table does not print a percent sign本次发布有 1 条成绩记录,含 1 条精确值(可含历史记录)