
llmfit
β Text source-checked Nov 20, 2026A terminal tool that right-sizes LLM models to your system's RAM, CPU, and GPU.
Source-checked against the vendor's site Β· rebuilt from the vendor's own pagesProfile updated Link checked How we rate
llmfit inspects the CPU, system RAM, GPUs, VRAM, and accelerator configuration to recommend models across popular quantizations. The vendor says that llmfit detects the hardware including RAM, CPU, GPU/VRAM, and backend, then scores every model in its catalog across four dimensions: memory fit, estimated speed, quality, and context. llmfit ships with an interactive TUI as the default interface and a classic CLI mode.
llmfit supports multi-GPU setups, MoE architectures, dynamic quantization selection, speed estimation, and local runtime providers including Ollama, llama.cpp, MLX, Docker Model Runner, and LM Studio. llmfit supports macOS (Apple Silicon & Intel), Linux (x86_64 & ARM64), and Windows (x86_64).
llmfit exposes standard HTTP JSON endpoints for integration into orchestrators, dashboards, and automated deployment pipelines. llmfit provides a multi-architecture Docker image supporting both interactive CLI/TUI and headless Web UI / API server modes.
According to llmfit: Llmfit inspects the CPU, System RAM, GPUs, VRAM, and accelerator configuration to recommend models across popular quantizations. Its pages list this strength: llmfit detects CPU cores, system RAM, available discrete/integrated GPUs, VRAM, and unified memory architecture for NVIDIA CUDA, Apple Silicon, AMD ROCm, and Intel OneAPI.
β The AI Tool Bible editorial team
Pros
- β llmfit detects CPU cores, system RAM, available discrete/integrated GPUs, VRAM, and unified memory architecture for NVIDIA CUDA, Apple Silicon, AMD ROCm, and Intel OneAPI.
- β The vendor says that users can choose between a lightweight, zero-dependency terminal interface or a feature-rich web dashboard.
- β llmfit provides a multi-architecture Docker image supporting both interactive CLI/TUI and headless Web UI / API server modes.
- β Users can download a model, serve it, measure real tok/s on their hardware, and contribute the results back to the project as a PR from the TUI.
Facts
- Platforms
- web, macos, linux, windows, cli (not source-checked)
- Pricing
- free (not source-checked)
- API available
- Yes
- Open source
- Yes (license: MIT) (not source-checked)
- GitHub stars
- 38,954 (checked Nov 13, 2026) (not source-checked)
- Last GitHub push
- Nov 12, 2026 (not source-checked)
- First commit
- Feb 2026 (not source-checked)
- Model used
- Multi-model (not source-checked)
- Vendor pages
- Changelog