# [datawhalechina/diy-llm](https://github.com/datawhalechina/diy-llm) Covers pre-training data, Tokenizer, Transformer, MoE,distributed training, Scaling Laws, inference & alignment .6 progressive code assignments for full-stack LLM learning | 涵盖预训练数据、分词器、Transformer、MoE、分布式训练、缩放定律、推理与对齐，6 项渐进代码作业，掌握 LLM 全栈知识 2.9k stars · 281 forks · Jupyter Notebook Quick update: star history is from the full report of Nov 1, 2026. Caution No license file. ## Why - GitHub found no license file, so the code is not clearly licensed for reuse. ([source 1](https://github.com/datawhalechina/diy-llm)) ### What would change this - A license file is added. - Partly checked: maintenance (Commit counts are estimated: the commit sample stops before the window start, so GitHub's totals were scaled by the sampled bot share.) ## Maintenance Partial Slowing **Human commits, 30 days** 36 **Human commits, 90 days** 60 **Human commits, 365 days** 771 **Bot share of commits, 365 days** 0% **Last commit** Nov 9, 2026 (2 days ago) **Last human commit** Nov 9, 2026 (2 days ago) **Authors, 90 days** 8 **Last release** Aug 17, 2026 (2 months ago) **Median PR merge time** 11 hours Responsiveness: of 5 recent issues, 100% got a maintainer reply. Median first response: 17 hours. - Commit counts are estimated: the commit sample stops before the window start, so GitHub's totals were scaled by the sampled bot share. Sources: [Commits](https://github.com/datawhalechina/diy-llm/commits), [Releases](https://github.com/datawhalechina/diy-llm/releases), [Merged pull requests](https://github.com/datawhalechina/diy-llm/pulls?q=is%3Apr+is%3Amerged), [Issues](https://github.com/datawhalechina/diy-llm/issues?q=is%3Aissue) ## People Complete **Owner** Organization **Commits sampled** 100 since Nov 12, 2025 **Authors for half of commits** 2 **Authors for 80% of commits** 3 **Contributors** 15 ### Top committers - xuhu0115: 39% of commits - 1iyouzhen: 21% of commits - kangkang-Adam: 21% of commits - jiangyinhe: 7% of commits - jiabaow326-cpu: 6% of commits - Shares come from the newest sampled commits, since the commit sample is capped. Sources: [Commits](https://github.com/datawhalechina/diy-llm/commits), [Contributors](https://github.com/datawhalechina/diy-llm/graphs/contributors) ## Stars Complete Organic, confidence medium Bars show stars gained each week. Red bars overlap a burst. ### Bursts - **Mar 6, 2026 to Mar 12, 2026**: about 43 more stars than usual, peak 13 in one day, a spike that tapers off. Outside explanation: a jump in forks. - **Apr 8, 2026 to Apr 9, 2026**: about 114 more stars than usual, peak 103 in one day, a spike that tapers off. Outside explanation: a jump in forks, a jump in issues and pull requests. - **Jul 30, 2026 to Jul 30, 2026**: about 28 more stars than usual, peak 31 in one day, a spike that tapers off. Outside explanation: a jump in issues and pull requests. - **Aug 12, 2026 to Aug 13, 2026**: about 33 more stars than usual, peak 29 in one day, a spike that tapers off. No outside explanation found, so this burst is unexplained. - **Oct 23, 2026 to Oct 31, 2026**: about 1k more stars than usual, peak 361 in one day, a shape we can't classify. Outside explanation: a jump in forks, a jump in issues and pull requests. Anyone can buy stars for any repo; bursts alone don't prove the owner did. - The star history looks organic. - Star history from the full report of Nov 1, 2026; this update did not re-read it. Sources: [How star history is analyzed](https://reporadar.ai/methodology#stars) ## Risk Complete ### Flags - **Warning:** GitHub finds no license for this repository, so default copyright rules apply and no reuse is granted. [Details](https://github.com/datawhalechina/diy-llm) ### License **Now** No license found **A year ago** Not known ### Packages No published packages found. Sources: [GitHub: datawhalechina/diy-llm](https://github.com/datawhalechina/diy-llm) ## Compare Weighing datawhalechina/diy-llm against an alternative? See both side by side. ## Badge ![RepoRadar: Caution](data:image/svg+xml,%3Csvg%20xmlns%3D%22http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%22%20width%3D%22112%22%20height%3D%2220%22%20role%3D%22img%22%20aria-label%3D%22reporadar%3A%20caution%22%3E%3Ctitle%3Ereporadar%3A%20caution%3C%2Ftitle%3E%3CclipPath%20id%3D%22r%22%3E%3Crect%20width%3D%22112%22%20height%3D%2220%22%20rx%3D%223%22%2F%3E%3C%2FclipPath%3E%3Cg%20clip-path%3D%22url(%23r)%22%3E%3Crect%20width%3D%2263%22%20height%3D%2220%22%20fill%3D%22%23555%22%2F%3E%3Crect%20x%3D%2263%22%20width%3D%2249%22%20height%3D%2220%22%20fill%3D%22%23dfa000%22%2F%3E%3C%2Fg%3E%3Cg%20fill%3D%22%23fff%22%20text-anchor%3D%22middle%22%20font-family%3D%22Verdana%2CGeneva%2CDejaVu%20Sans%2Csans-serif%22%20font-size%3D%2211%22%3E%3Ctext%20x%3D%2232%22%20y%3D%2214%22%3Ereporadar%3C%2Ftext%3E%3Ctext%20x%3D%2288%22%20y%3D%2214%22%3Ecaution%3C%2Ftext%3E%3C%2Fg%3E%3C%2Fsvg%3E) Add this to a README to show the current verdict: ``` [![RepoRadar](https://reporadar.ai/datawhalechina/diy-llm/badge.svg)](https://reporadar.ai/datawhalechina/diy-llm) ``` ## Follow Get told when this verdict changes. Add the feed to any feed reader: you get one entry per change, not one per refresh. ``` https://reporadar.ai/datawhalechina/diy-llm/feed.xml ``` [Open the feed](https://reporadar.ai/datawhalechina/diy-llm/feed.xml) ## More repos rated Caution [See all](https://reporadar.ai/-/reports/caution) - [jnsahaj/lumen](https://reporadar.ai/jnsahaj/lumen) 2.9k stars - [datawhalechina/deepagents-in-action](https://reporadar.ai/datawhalechina/deepagents-in-action) 2.9k stars - [iFurySt/open-computer-use](https://reporadar.ai/iFurySt/open-computer-use) 2.9k stars - [NeptuneHub/AudioMuse-AI](https://reporadar.ai/NeptuneHub/AudioMuse-AI) 2.9k stars - [webosbrew/dev-manager-desktop](https://reporadar.ai/webosbrew/dev-manager-desktop) 2.9k stars - [sahibzada-allahyar/Formal-Verification-RSI](https://reporadar.ai/sahibzada-allahyar/Formal-Verification-RSI) 2.9k stars ## Earlier verdicts - Nov 12, 2026: Caution - Nov 11, 2026: Caution - Nov 10, 2026: Caution - Nov 9, 2026: Caution - Nov 8, 2026: Caution - Nov 7, 2026: Caution - Nov 6, 2026: Caution - Nov 5, 2026: Caution - Nov 4, 2026: Caution - Nov 3, 2026: Caution Computed 10 hours ago · method 2026.10.2 Source: https://reporadar.ai/datawhalechina/diy-llm