llmfit
Share on:
Enter Your Mastodon Instance
Copy the link below to share to Mastodon
https://terminaltrove.com/llmfit/
A terminal tool that finds what LLM models can run on your system's RAM, CPU, and GPU.
Script Preview
Review this script before executing in your terminal.
llmfit is a terminal tool that checks which local LLMs can fit on your machine and ranks how well it can run on your hardware.
It automatically detects your system specifications such as RAM, VRAM, CPU, GPU and inference runtime and then scores models by quality, speed, fit, and required memory consumption. You can compare over hundreds of models against your machines requirements and it allows you to filter more than 20 different properties, such as parameter count, the model architecture, quantization, context size, tokens per second, and how well the model fits on your machine.
You can search the model name in the search bar and you can filter by model capability (reasoning, tool use, chat, coding, multimodal, embedding) and it detects which runtime to run the model under (runtimes such as vLLM, MLX, llama.cpp, Docker Model Runner, and LM Studio.)
This tool is very useful to AI researchers, developers, engineers, and AI enthusiasts who are interested in self-hosted LLM inference engineering who want to pick the best model for their machine depending on their laptop, desktop, or GPU box specs in the terminal.
