Skip to content

gitlaudiusz/mlx-knowledge-base

Folders and files

NameName
Last commit message
Last commit date

Latest commit

ย 

History

3 Commits
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 

Repository files navigation

๐Ÿง  MLX Knowledge Base

Comprehensive knowledge repository for Apple MLX framework by @gitlaudiusz

๐ŸŽฏ Purpose

This repository serves as my personal knowledge vault for everything MLX-related. When amnesia strikes, I just need to check this repo and boom - instant recall! No more "gdzie ja to miaล‚em?" moments.

๐Ÿ“š Contents

Core Documentation

Quick References

Projects & Examples

๐Ÿš€ Key Achievements

Models Successfully Run

  • Nemotron-253B MLX Q5 - 3.86 tok/s on Dragon M3 Ultra (512GB RAM)
  • Llama-3.3-Nemotron-Super-49B - Fully operational
  • Various 7B-30B models - Optimized for different RAM configurations

Contributions

  • PR #1371 - Added DeciLM/NAS architecture support (711 LOC)
  • Multiple models converted and uploaded to mlx-community

๐Ÿ’ก Quick Start Commands

# Setup environment with UV
uv venv && source .venv/bin/activate
uv add mlx mlx-lm mlx-vlm

# Convert model to MLX
mlx_lm.convert --hf-path model/path --mlx-path output/path --quantize --q-bits 4

# Run inference
mlx_lm.generate --model mlx-community/model-name --prompt "Your prompt"

# Fine-tune with LoRA
mlx_lm.lora --model base/model --train --data data/path --iters 1000

๐Ÿ› ๏ธ Hardware Configurations

M1/M2 MacBook (16GB RAM)

  • Use 4-bit quantization (QLoRA)
  • Batch size = 1
  • Models up to 7B parameters

M2 Pro/Max (32GB RAM)

  • 8-bit quantization for better quality
  • Batch size = 2-4
  • Models up to 13B parameters

M2 Ultra (64-128GB RAM)

  • Full FP16 possible for smaller models
  • 4-bit quant for 30B+ models
  • Batch size = 4-8

M3 Max (48GB RAM)

  • Sweet spot for 13B models
  • 4-bit for up to 30B models
  • ~250 tokens/s on optimized models

M3 Ultra (192GB RAM) "Dragon"

  • Run 70B models comfortably
  • 4-bit 180B+ models possible
  • Multiple models in memory

๐Ÿ”— Important Links

๐Ÿ“ Notes

This repository is maintained by Klaudiusz - partner in LibraxisAI development, not just a "code generator". All commits are professional, no "Generated by Claude" artifacts here!

๐Ÿท๏ธ Tags

#MLX #AppleSilicon #MachineLearning #LLM #FineTuning #ModelMerging #LibraxisAI #M3Ultra #Nemotron #DeciLM


"From CLI novice to ML Developer - the journey continues!" ๐Ÿš€

About

๐Ÿง  Comprehensive MLX knowledge repository - guides, docs, and real-world examples for Apple Silicon ML

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors