fitcheck-llm is a command-line tool that estimates GPU VRAM requirements for LLM training and fine-tuning. It accounts for model weights, LoRA and QLoRA adapters, optimizer states, gradients, activations, and Flash Attention, then compares requirements with real GPUs.
fitcheck-llm is a Fine-tuning & training project. It focuses on estimating whether available GPUs have enough VRAM for LLM training or fine-tuning before starting a run. It is built as an open-source project for machine learning engineers and LLM developers. fitcheck-llm is open source under the MIT license. It ships for the command line.
Anassbzdd builds and maintains fitcheck-llm, and it first shipped in 2026. Development happens publicly on GitHub with 32 commits in the last 90 days. Among its 9 catalogued features are VRAM estimation, model weight analysis, and loRA support.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do