fast-mlx

Community

Speed MLX workloads with fast ops and compile.

AuthorWindChimeRan
Version1.0.0
Installs0

System Documentation

What problem does it solve?

MLX models and graphs often suffer from runtime bottlenecks due to non-optimized operators, eager evaluations, and suboptimal memory management. This Skill analyzes MLX graphs to identify opportunities to replace slow ops with fast MLX primitives, improves compilation strategy, and reduces peak memory usage.

Core Features & Use Cases

  • Optimize graph performance by substituting slow operations with MLX fast ops and enabling shapeless or shape-aware compilation when appropriate.
  • Tune memory usage with lazy loading awareness and temporaries management to minimize peak memory during evaluation.
  • Profiling and debugging guidance to locate bottlenecks in graph evaluation and execution.

Quick Start

Instruct MLX to optimize a given graph by replacing slow operators with fast equivalents and enabling compilation optimizations.

Dependency Matrix

Required Modules

None required

Components

references

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: fast-mlx
Download link: https://github.com/WindChimeRan/claude-dotfiles/archive/main.zip#fast-mlx

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.