--- title: "self-llm vs LLMs-from-scratch" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/datawhalechina-self-llm-vs-rasbt-llms-from-scratch" tools: ["datawhalechina-self-llm", "rasbt-llms-from-scratch"] --- # self-llm vs LLMs-from-scratch Neutral, constraint-first comparison with live GitHub stats. | | [self-llm](/tools/datawhalechina-self-llm.md) | [LLMs-from-scratch](/tools/rasbt-llms-from-scratch.md) | | --- | --- | --- | | Tagline | 针对中国用户的开源大模型教程 | Implement a ChatGPT-like LLM in PyTorch from scratch, step by step | | Stars | 31,200 | 98,748 | | Forks | 3,047 | 15,153 | | Open issues | 158 | 4 | | Language | Jupyter Notebook | Jupyter Notebook | | Adopt for | Self-LLM is a comprehensive tutorial repository for deploying and fine-tuning large language models (LLMs) tailored for Chinese users, focusing on accessibility through Linux-based configurations. With extensive support, | LLMs-from-scratch is a repository that offers detailed, step-by-step guidance on developing, pretraining, and finetuning GPT-like large language models using PyTorch. The codebase complements a book dedicated to building | | Persona | - | - | | Runtime | - | - | | License | Apache-2.0 | Other | | Categories | LLM Frameworks, Model Training, Inference & Serving | LLM Frameworks, Model Training | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [self-llm](/tools/datawhalechina-self-llm.md) | [LLMs-from-scratch](/tools/rasbt-llms-from-scratch.md) | | --- | --- | --- | | Maintenance | Active (82%) | Steady (60%) | | Days since push | 21d | 35d | | Open issues (now) | 158 | 4 | | Owner type | Organization | User | | Security scan | No lockfile | 34 low (34 low) | | Full report | [trust report](/tools/datawhalechina-self-llm/trust.md) | [trust report](/tools/rasbt-llms-from-scratch/trust.md) | **Typed relationship:** self-llm _(related)_ LLMs-from-scratch ## Decision facts: self-llm - **Adopt for:** Self-LLM is a comprehensive tutorial repository for deploying and fine-tuning large language models (LLMs) tailored for Chinese users, focusing on accessibility through Linux-based configurations. With extensive support, ## Decision facts: LLMs-from-scratch - **Requirements:** Min 8 GB RAM; The repository includes comprehensive documentation that can be used alongside the book 'Build a Large Language Model (From Scratch)' for additional context and - **Adopt for:** LLMs-from-scratch is a repository that offers detailed, step-by-step guidance on developing, pretraining, and finetuning GPT-like large language models using PyTorch. The codebase complements a book dedicated to building ## Choose when ### Choose self-llm if… - License: self-llm is Apache-2.0, LLMs-from-scratch is Other. - Graph edge: self-llm is a typed related of LLMs-from-scratch - see the relationship row above. - Tags unique to self-llm: qwen, lora, deployment, micro-tuning. - Also covers Inference & Serving. - You are located in China and require detailed, locale-specific guidance to deploy LLMs. ### Choose LLMs-from-scratch if… - License: LLMs-from-scratch is Other, self-llm is Apache-2.0. - Requirements: Min 8 GB RAM; The repository includes comprehensive documentation that can be used alongside the book 'Build a Large Language Model (From Scratch)' for additional context and. - Graph edge: LLMs-from-scratch is a typed related of self-llm - see the relationship row above. - Tags unique to LLMs-from-scratch: deep-learning, ai, artificial-intelligence, instruction-tuning. - When you need detailed, step-by-step explanations and examples for constructing an LLM from scratch with PyTorch. ## When NOT to use self-llm - Your primary platform is Windows-based, as the detailed deployment instructions and configurations are Linux-oriented. - You require a more graphical user interface (GUI)-based approach rather than command-line interaction to deploy LLMs, since this resource emphasizes terminal-based configurations. ## When NOT to use LLMs-from-scratch - When you are looking for a quick setup or already have familiarity with LLMs as the repository emphasizes building from scratch, which can be time-consuming. - If your primary goal is production-scale deployment rather than educational understanding, as this tool focuses more on learning through thoroughness rather than speed and optimization. - For users who prefer not to use specific frameworks like PyTorch and are interested in developing models with other libraries. ## Common questions ### What is the difference between self-llm and LLMs-from-scratch? self-llm: 针对中国用户的开源大模型教程. LLMs-from-scratch: Implement a ChatGPT-like LLM in PyTorch from scratch, step by step. See the comparison table for live GitHub stats and shared categories. ### When should I choose self-llm over LLMs-from-scratch? Choose self-llm over LLMs-from-scratch when License: self-llm is Apache-2.0, LLMs-from-scratch is Other; Graph edge: self-llm is a typed related of LLMs-from-scratch - see the relationship row above; Tags unique to self-llm: qwen, lora, deployment, micro-tuning; Also covers Inference & Serving; You are located in China and require detailed, locale-specific guidance to deploy LLMs. ### When should I choose LLMs-from-scratch over self-llm? Choose LLMs-from-scratch over self-llm when License: LLMs-from-scratch is Other, self-llm is Apache-2.0; Requirements: Min 8 GB RAM; The repository includes comprehensive documentation that can be used alongside the book 'Build a Large Language Model (From Scratch)' for additional context and; Graph edge: LLMs-from-scratch is a typed related of self-llm - see the relationship row above; Tags unique to LLMs-from-scratch: deep-learning, ai, artificial-intelligence, instruction-tuning; When you need detailed, step-by-step explanations and examples for constructing an LLM from scratch with PyTorch. ### When should I avoid self-llm? Your primary platform is Windows-based, as the detailed deployment instructions and configurations are Linux-oriented. You require a more graphical user interface (GUI)-based approach rather than command-line interaction to deploy LLMs, since this resource emphasizes terminal-based configurations. ### When should I avoid LLMs-from-scratch? When you are looking for a quick setup or already have familiarity with LLMs as the repository emphasizes building from scratch, which can be time-consuming. If your primary goal is production-scale deployment rather than educational understanding, as this tool focuses more on learning through thoroughness rather than speed and optimization. For users who prefer not to use specific frameworks like PyTorch and are interested in developing models with other libraries. ### Is self-llm or LLMs-from-scratch more popular on GitHub? LLMs-from-scratch has more GitHub stars (98,748 vs 31,200). Stars measure visibility, not whether either tool fits your constraints. ### Are self-llm and LLMs-from-scratch open source? Yes - both are open-source projects on GitHub (self-llm: Apache-2.0, LLMs-from-scratch: Other). ### Where can I find alternatives to self-llm or LLMs-from-scratch? GraphCanon lists graph-backed alternatives at /tools/datawhalechina-self-llm/alternatives and /tools/rasbt-llms-from-scratch/alternatives (/tools/datawhalechina-self-llm/alternatives.md, /tools/rasbt-llms-from-scratch/alternatives.md), ranked by typed relationship edges rather than popularity votes. ### Is there a machine-readable version of this comparison? Yes. The markdown twin at /compare/datawhalechina-self-llm-vs-rasbt-llms-from-scratch.md mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, self-llm or LLMs-from-scratch? self-llm: Active. LLMs-from-scratch: Steady. Compare maintenance labels, days since push, and release cadence in the trust section below - stars alone do not measure maintenance. ### Where are the full trust reports for self-llm and LLMs-from-scratch? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: self-llm: /tools/datawhalechina-self-llm/trust; LLMs-from-scratch: /tools/rasbt-llms-from-scratch/trust. --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=datawhalechina-self-llm`](/api/graphcanon/graph?tool=datawhalechina-self-llm) - LLM index: [/llms.txt](/llms.txt) - Full corpus: [/llms-full.txt](/llms-full.txt) _GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_