https://github.com/johnmai-dev/NotebookMLX

📋 NotebookMLX - An Open Source version of NotebookLM (Ported NotebookLlama)
https://github.com/johnmai-dev/NotebookMLX

ai llama mlx notebookllama notebooklm

Last synced: 6 months ago
JSON representation

📋 NotebookMLX - An Open Source version of NotebookLM (Ported NotebookLlama)

• Host: GitHub
• URL: https://github.com/johnmai-dev/NotebookMLX
• Owner: johnmai-dev
• License: mit
• Created: 2024-10-29T13:53:13.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2025-03-03T16:31:37.000Z (7 months ago)
• Last Synced: 2025-03-28T04:15:29.899Z (6 months ago)
• Topics: ai, llama, mlx, notebookllama, notebooklm
• Language: Jupyter Notebook
• Homepage:
• Size: 92.6 MB
• Stars: 267
• Watchers: 4
• Forks: 19
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# NotebookMLX

> [meta-llama/NotebookLlama](https://github.com/meta-llama/llama-recipes/tree/main/recipes/quickstart/NotebookLlama)

I ported [NotebookLlama](https://github.com/meta-llama/llama-recipes/tree/main/recipes/quickstart/NotebookLlama) and

implemented it with [MLX](https://github.com/ml-explore/mlx) 🔥

It uses [mlx-community/Qwen2.5-1.5B-Instruct-bf16](https://huggingface.co/mlx-community/Qwen2.5-1.5B-Instruct-4bit) for

pre-processing the

PDF, [mlx-community/Qwen2.5-14B-Instruct-4bit](https://huggingface.co/mlx-community/Qwen2.5-14B-Instruct-4bit) for

creating

transcripts, [mlx-community/Qwen2.5-7B-Instruct-4bit](https://huggingface.co/mlx-community/Qwen2.5-7B-Instruct-4bit) for

rewrites, and [lucasnewman/f5-tts-mlx](https://huggingface.co/lucasnewman/f5-tts-mlx) for Text-to-Speech ⚡

> Citing the NotebookLlama outline.

> ![Outline.jpg](resources/Outline.jpg)

[Step 1](Step-1-PDF-Pre-Processing-Logic.ipynb): Pre-process PDF:

Use [mlx-community/Qwen2.5-1.5B-Instruct-bf16](https://huggingface.co/mlx-community/Qwen2.5-1.5B-Instruct-4bit) to

pre-process the PDF and save it in a .txt file.

[Step 2](Step-2-Transcript-Writer.ipynb): Transcript Writer:

Use [mlx-community/Qwen2.5-14B-Instruct-4bit](https://huggingface.co/mlx-community/Qwen2.5-14B-Instruct-4bit) to write a

podcast transcript from the text.

[Step 3](Step-3-Re-Writer.ipynb): Dramatic Re-Writer: Use

the [mlx-community/Qwen2.5-7B-Instruct-4bit](https://huggingface.co/mlx-community/Qwen2.5-7B-Instruct-4bit) model to

make the transcript more dramatic.

[Step 4](Step-4-TTS-Workflow.ipynb): Text-To-Speech Workflow:

Use [lucasnewman/f5-tts-mlx](https://huggingface.co/lucasnewman/f5-tts-mlx) to generate

a conversational podcast.

## The podcast audio

https://github.com/user-attachments/assets/257e4822-1af0-479a-94a4-a61c7abe434b

## Star History