Repolode
Submit
zkywsg

zkywsg/Daily-LLM

A visual, timeline-based learning roadmap that traces deep learning and LLM architecture families from the 1986 RNN to 2025 reasoning models.

View on GitHub
Stars
869
Forks
158
Open issues
0
Last commit
last month

What is it?

Daily-LLM is a bilingual learning path for deep learning and large language models. It organizes the field into 19 architecture families, from CNNs and RNN/LSTM through Transformers, BERT and GPT, diffusion models, LoRA fine-tuning, RLHF, mixture-of-experts, RAG and agents, up to reasoning models such as o1 and R1. Each family has its own directory, and a yearly index is generated in TIMELINE.md.

Beyond the written modules, the repository includes a local interactive web page with timeline and family views, and cross-family projects under a projects directory. Foundational topics such as activations, backpropagation, optimizers, normalization and attention are collected in a separate foundations section.

The repository is useful for developers who want to understand how key ideas connected over time, or who are preparing for algorithm interviews and want a structured overview of the main model families.

Key features

  • 01Timeline of 19 model families from 1986 to 2025, each with its own directory
  • 02Local interactive web page with timeline and family browsing views
  • 03Cross-family hands-on projects in a dedicated projects folder
  • 04Foundations section covering activations, backprop, optimizers and attention
  • 05Bilingual documentation in English and Simplified Chinese

Use cases

  • Getting an organized overview of how deep learning architectures evolved
  • Preparing for machine learning and LLM algorithm interviews
  • Finding starter material on specific families such as diffusion models or RAG

Quick start

  1. 1

    Clone the repository

    git clone https://github.com/zkywsg/Daily-LLM.git
  2. 2

    Enter the project directory

    cd Daily-LLM
  3. 3

    Install dependencies from requirements.txt

    pip install -r requirements.txt

Things to know

  • This is a learning roadmap and study material, not a software library or application.

In collections

Natural language processing tools & libraries

Topics

#deep-learning#llm#machine-learning#transformer#tutorial#nlp#pytorch#cv#leetcode#leetcode-python

FAQ

What is Daily-LLM?
A visual, timeline-based learning roadmap that traces deep learning and LLM architecture families from the 1986 RNN to 2025 reasoning models. Daily-LLM is a bilingual learning path for deep learning and large language models. It organizes the field into 19 architecture families, from CNNs and RNN/LSTM through Transformers, BERT and GPT, diffusion models, LoRA fine-tuning, RLHF, mixture-of-experts, RAG and agents, up to reasoning models such as o1 and R1. Each family has its own directory, and a yearly index is generated in TIMELINE.md.
Is Daily-LLM free to use? What license does it have?
Daily-LLM is open source under the MIT license. Check the license terms before using it in commercial products.
Is Daily-LLM still maintained?
Updated within 6 months. The last commit to Daily-LLM was on Sep 25, 2026.
What can I use Daily-LLM for?
Getting an organized overview of how deep learning architectures evolved Preparing for machine learning and LLM algorithm interviews Finding starter material on specific families such as diffusion models or RAG
How do I install Daily-LLM?
Clone the repository: git clone https://github.com/zkywsg/Daily-LLM.git Enter the project directory: cd Daily-LLM Install dependencies from requirements.txt: pip install -r requirements.txt
Where can I get Daily-LLM?
The source code and installation instructions are on GitHub: https://github.com/zkywsg/Daily-LLM

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects