Overview

Unlock the full potential of your hardware with llmfit, the ultimate command-line tool for discovering and running Large Language Models (LLMs). llmfit intelligently analyzes your system's RAM, CPU, and GPU, then meticulously scores hundreds of models and providers based on quality, speed, fit, and context. Say goodbye to guesswork and hello to effortless LLM deployment, tailored precisely to your machine.

Key Features

• Hardware-Aware Model Selection: Automatically detects your system's specifications to recommend LLMs that will run optimally.
• Comprehensive Model Catalog: Access and evaluate a vast collection of LLMs from numerous providers.
• Intelligent Scoring System: Models are ranked across four critical dimensions: memory fit, estimated speed, quality, and context window.
• Interactive TUI & Classic CLI: Seamlessly switch between a user-friendly terminal interface for exploration and a powerful command-line mode for scripting and automation.
• Multi-Provider Support: Integrates with popular local LLM runtimes including Ollama, llama.cpp, MLX, Docker Model Runner, and LM Studio.
• Advanced Features: Supports multi-GPU setups, MoE architectures, dynamic quantization, and real-time speed estimation.
• Verifiable Benchmarking: Measure and contribute real-world performance metrics directly from your hardware, enhancing community-driven accuracy.

Typical Use Cases

• Developers & Researchers: Quickly identify the best-performing LLMs for your specific hardware without manual configuration or trial-and-error.
• ML Engineers: Integrate llmfit into your CI/CD pipelines for automated LLM selection and deployment based on target hardware profiles.
• Hobbyists & Enthusiasts: Effortlessly find LLMs that run smoothly on your personal machine, whether for creative writing, coding assistance, or experimentation.
• System Administrators: Deploy LLMs across diverse hardware fleets with confidence, ensuring optimal performance and resource utilization.