AlexsJones/llmfitPublic

Hundreds of models & providers. One command to find what runs on your hardware.

AI summary: A terminal user interface for benchmarking large language models directly on local hardware configurations.

Stars
39.2K
+129 today
Forks
2.5K
Watchers
115
Open issues
36
Open PRs
29
Contributors
~182
Commits
1.4K
Branches
158

RustMITCreated Feb 15, 2026Last push 2d agoLatest release v1.1.20+185 stars this week+962 this month

Quick answers

What is llmfit?
A terminal user interface for benchmarking large language models directly on local hardware configurations.
What does llmfit do?
LLMFit is a highly localized benchmarking tool that evaluates the real-world performance of large language models on specific hardware. It bypasses theoretical estimates by actually downloading, serving, and measuring true tokens-per-second directly on the user's machine. The entire process is managed through a clean Terminal User Interface (TUI), eliminating complex command-line configurations. It provides actionable performance metrics that users can optionally share to build an accurate, community-driven hardware database without requiring third-party accounts. It supports dynamic quantization selection and directly interacts with local runtimes like Ollama and llama.cpp.
Who is llmfit for?
AI developers, hardware enthusiasts, and system administrators running large language models locally who require accurate metrics. The tool offers clear performance insights for these specific hardware setups.
How do I get started with llmfit?
cargo install llmfit
How popular is llmfit on GitHub?
AlexsJones/llmfit has 39,153 stars and 2,536 forks on GitHub, and gained 185 stars in the last 7 days.
What license does llmfit use?
AlexsJones/llmfit is released under the MIT license.

Star history

since Jul 28, 2026
010K20K30KJul 2026Sep 2026Oct 2026Nov 2026
39.2K stars as of Nov 20, 2026. Measured daily since Jul 28, 2026; GitHub no longer exposes earlier star timestamps.

Update history

3 recorded
  • Nov 13, 2026New release v1.1.20 (previously v1.1.18).
  • Nov 2, 2026New release v1.1.18 (previously v1.1.17).
  • Oct 13, 2026New release v1.1.17 (previously v1.1.16).

Contribution activity

commits per day, last 52 weeks
NovDecJanFebMarAprMayJunJulAugSepOctNovMonWedFri2025-11-16: 0 commits2025-11-17: 0 commits2025-11-18: 0 commits2025-11-19: 0 commits2025-11-20: 0 commits2025-11-21: 0 commits2025-11-22: 0 commits2025-11-23: 0 commits2025-11-24: 0 commits2025-11-25: 0 commits2025-11-26: 0 commits2025-11-27: 0 commits2025-11-28: 0 commits2025-11-29: 0 commits2025-11-30: 0 commits2025-12-01: 0 commits2025-12-02: 0 commits2025-12-03: 0 commits2025-12-04: 0 commits2025-12-05: 0 commits2025-12-06: 0 commits2025-12-07: 0 commits2025-12-08: 0 commits2025-12-09: 0 commits2025-12-10: 0 commits2025-12-11: 0 commits2025-12-12: 0 commits2025-12-13: 0 commits2025-12-14: 0 commits2025-12-15: 0 commits2025-12-16: 0 commits2025-12-17: 0 commits2025-12-18: 0 commits2025-12-19: 0 commits2025-12-20: 0 commits2025-12-21: 0 commits2025-12-22: 0 commits2025-12-23: 0 commits2025-12-24: 0 commits2025-12-25: 0 commits2025-12-26: 0 commits2025-12-27: 0 commits2025-12-28: 0 commits2025-12-29: 0 commits2025-12-30: 0 commits2025-12-31: 0 commits2026-01-01: 0 commits2026-01-02: 0 commits2026-01-03: 0 commits2026-01-04: 0 commits2026-01-05: 0 commits2026-01-06: 0 commits2026-01-07: 0 commits2026-01-08: 0 commits2026-01-09: 0 commits2026-01-10: 0 commits2026-01-11: 0 commits2026-01-12: 0 commits2026-01-13: 0 commits2026-01-14: 0 commits2026-01-15: 0 commits2026-01-16: 0 commits2026-01-17: 0 commits2026-01-18: 0 commits2026-01-19: 0 commits2026-01-20: 0 commits2026-01-21: 0 commits2026-01-22: 0 commits2026-01-23: 0 commits2026-01-24: 0 commits2026-01-25: 0 commits2026-01-26: 0 commits2026-01-27: 0 commits2026-01-28: 0 commits2026-01-29: 0 commits2026-01-30: 0 commits2026-01-31: 0 commits2026-02-01: 0 commits2026-02-02: 0 commits2026-02-03: 0 commits2026-02-04: 0 commits2026-02-05: 0 commits2026-02-06: 0 commits2026-02-07: 0 commits2026-02-08: 0 commits2026-02-09: 0 commits2026-02-10: 0 commits2026-02-11: 0 commits2026-02-12: 0 commits2026-02-13: 0 commits2026-02-14: 0 commits2026-02-15: 22 commits2026-02-16: 20 commits2026-02-17: 5 commits2026-02-18: 14 commits2026-02-19: 9 commits2026-02-20: 7 commits2026-02-21: 18 commits2026-02-22: 7 commits2026-02-23: 18 commits2026-02-24: 8 commits2026-02-25: 2 commits2026-02-26: 13 commits2026-02-27: 9 commits2026-02-28: 12 commits2026-03-01: 14 commits2026-03-02: 13 commits2026-03-03: 16 commits2026-03-04: 15 commits2026-03-05: 6 commits2026-03-06: 4 commits2026-03-07: 2 commits2026-03-08: 3 commits2026-03-09: 5 commits2026-03-10: 5 commits2026-03-11: 2 commits2026-03-12: 11 commits2026-03-13: 6 commits2026-03-14: 4 commits2026-03-15: 2 commits2026-03-16: 10 commits2026-03-17: 4 commits2026-03-18: 23 commits2026-03-19: 4 commits2026-03-20: 2 commits2026-03-21: 3 commits2026-03-22: 8 commits2026-03-23: 7 commits2026-03-24: 0 commits2026-03-25: 0 commits2026-03-26: 1 commit2026-03-27: 2 commits2026-03-28: 0 commits2026-03-29: 1 commit2026-03-30: 4 commits2026-03-31: 1 commit2026-04-01: 5 commits2026-04-02: 1 commit2026-04-03: 1 commit2026-04-04: 12 commits2026-04-05: 4 commits2026-04-06: 7 commits2026-04-07: 0 commits2026-04-08: 1 commit2026-04-09: 6 commits2026-04-10: 0 commits2026-04-11: 5 commits2026-04-12: 2 commits2026-04-13: 8 commits2026-04-14: 2 commits2026-04-15: 2 commits2026-04-16: 0 commits2026-04-17: 3 commits2026-04-18: 6 commits2026-04-19: 3 commits2026-04-20: 4 commits2026-04-21: 5 commits2026-04-22: 11 commits2026-04-23: 6 commits2026-04-24: 4 commits2026-04-25: 5 commits2026-04-26: 1 commit2026-04-27: 12 commits2026-04-28: 1 commit2026-04-29: 5 commits2026-04-30: 2 commits2026-05-01: 1 commit2026-05-02: 1 commit2026-05-03: 4 commits2026-05-04: 4 commits2026-05-05: 2 commits2026-05-06: 4 commits2026-05-07: 0 commits2026-05-08: 0 commits2026-05-09: 0 commits2026-05-10: 3 commits2026-05-11: 0 commits2026-05-12: 4 commits2026-05-13: 1 commit2026-05-14: 0 commits2026-05-15: 0 commits2026-05-16: 0 commits2026-05-17: 3 commits2026-05-18: 4 commits2026-05-19: 3 commits2026-05-20: 4 commits2026-05-21: 2 commits2026-05-22: 3 commits2026-05-23: 2 commits2026-05-24: 2 commits2026-05-25: 0 commits2026-05-26: 6 commits2026-05-27: 3 commits2026-05-28: 0 commits2026-05-29: 1 commit2026-05-30: 0 commits2026-05-31: 0 commits2026-06-01: 8 commits2026-06-02: 1 commit2026-06-03: 3 commits2026-06-04: 0 commits2026-06-05: 3 commits2026-06-06: 0 commits2026-06-07: 4 commits2026-06-08: 3 commits2026-06-09: 8 commits2026-06-10: 10 commits2026-06-11: 0 commits2026-06-12: 1 commit2026-06-13: 1 commit2026-06-14: 1 commit2026-06-15: 2 commits2026-06-16: 0 commits2026-06-17: 4 commits2026-06-18: 0 commits2026-06-19: 6 commits2026-06-20: 0 commits2026-06-21: 0 commits2026-06-22: 4 commits2026-06-23: 12 commits2026-06-24: 1 commit2026-06-25: 1 commit2026-06-26: 0 commits2026-06-27: 1 commit2026-06-28: 4 commits2026-06-29: 1 commit2026-06-30: 1 commit2026-07-01: 0 commits2026-07-02: 12 commits2026-07-03: 19 commits2026-07-04: 4 commits2026-07-05: 7 commits2026-07-06: 4 commits2026-07-07: 2 commits2026-07-08: 1 commit2026-07-09: 14 commits2026-07-10: 30 commits2026-07-11: 2 commits2026-07-12: 0 commits2026-07-13: 11 commits2026-07-14: 6 commits2026-07-15: 1 commit2026-07-16: 4 commits2026-07-17: 4 commits2026-07-18: 2 commits2026-07-19: 2 commits2026-07-20: 12 commits2026-07-21: 12 commits2026-07-22: 2 commits2026-07-23: 0 commits2026-07-24: 0 commits2026-07-25: 0 commits2026-07-26: 0 commits2026-07-27: 0 commits2026-07-28: 0 commits2026-07-29: 0 commits2026-07-30: 3 commits2026-07-31: 2 commits2026-08-01: 0 commits2026-08-02: 3 commits2026-08-03: 7 commits2026-08-04: 3 commits2026-08-05: 12 commits2026-08-06: 0 commits2026-08-07: 0 commits2026-08-08: 0 commits2026-08-09: 14 commits2026-08-10: 5 commits2026-08-11: 2 commits2026-08-12: 4 commits2026-08-13: 3 commits2026-08-14: 1 commit2026-08-15: 0 commits2026-08-16: 0 commits2026-08-17: 15 commits2026-08-18: 7 commits2026-08-19: 2 commits2026-08-20: 1 commit2026-08-21: 1 commit2026-08-22: 2 commits2026-08-23: 2 commits2026-08-24: 6 commits2026-08-25: 3 commits2026-08-26: 0 commits2026-08-27: 5 commits2026-08-28: 16 commits2026-08-29: 0 commits2026-08-30: 6 commits2026-08-31: 3 commits2026-09-01: 9 commits2026-09-02: 2 commits2026-09-03: 9 commits2026-09-04: 0 commits2026-09-05: 0 commits2026-09-06: 1 commit2026-09-07: 1 commit2026-09-08: 0 commits2026-09-09: 4 commits2026-09-10: 5 commits2026-09-11: 2 commits2026-09-12: 0 commits2026-09-13: 1 commit2026-09-14: 12 commits2026-09-15: 0 commits2026-09-16: 2 commits2026-09-17: 0 commits2026-09-18: 1 commit2026-09-19: 19 commits2026-09-20: 3 commits2026-09-21: 1 commit2026-09-22: 0 commits2026-09-23: 0 commits2026-09-24: 1 commit2026-09-25: 0 commits2026-09-26: 0 commits2026-09-27: 0 commits2026-09-28: 1 commit2026-09-29: 3 commits2026-09-30: 0 commits2026-10-01: 1 commit2026-10-02: 0 commits2026-10-03: 4 commits2026-10-04: 4 commits2026-10-05: 0 commits2026-10-06: 0 commits2026-10-07: 5 commits2026-10-08: 8 commits2026-10-09: 8 commits2026-10-10: 1 commit2026-10-11: 0 commits2026-10-12: 4 commits2026-10-13: 1 commit2026-10-14: 0 commits2026-10-15: 15 commits2026-10-16: 1 commit2026-10-17: 1 commit2026-10-18: 1 commit2026-10-19: 2 commits2026-10-20: 4 commits2026-10-21: 1 commit2026-10-22: 0 commits2026-10-23: 0 commits2026-10-24: 2 commits2026-10-25: 5 commits2026-10-26: 11 commits2026-10-27: 13 commits2026-10-28: 8 commits2026-10-29: 2 commits2026-10-30: 12 commits2026-10-31: 1 commit2026-11-01: 0 commits2026-11-02: 1 commit2026-11-03: 28 commits2026-11-04: 0 commits2026-11-05: 1 commit2026-11-06: 0 commits2026-11-07: 1 commit2026-11-08: 1 commit2026-11-09: 4 commits2026-11-10: 0 commits2026-11-11: 2 commits2026-11-12: 1 commit2026-11-13: 0 commits2026-11-14: 0 commits
1,113 commits in the last yearLessMore

Signals and awards

derived from tracked data
  • Widely adopted

    39,153 stars

  • Very active

    1,113 commits in 52 weeks

  • Community-driven

    ~182 contributors

  • Permissive license

    MIT

  • Continuous integration

    Automated checks passing

  • Repeat trending

    11 trending appearances

What llmfit does

LLMFit is a highly localized benchmarking tool that evaluates the real-world performance of large language models on specific hardware. It bypasses theoretical estimates by actually downloading, serving, and measuring true tokens-per-second directly on the user's machine. The entire process is managed through a clean Terminal User Interface (TUI), eliminating complex command-line configurations. It provides actionable performance metrics that users can optionally share to build an accurate, community-driven hardware database without requiring third-party accounts. It supports dynamic quantization selection and directly interacts with local runtimes like Ollama and llama.cpp.

AI developers, hardware enthusiasts, and system administrators running large language models locally who require accurate metrics. The tool offers clear performance insights for these specific hardware setups.

  • Hardware-Specific Benchmarking: Measures actual tokens-per-second on the exact machine rather than relying on generalized estimates.
  • Integrated Terminal Interface: Manages model downloads, test execution, and result viewing entirely from a clean TUI.
  • Direct PR Submissions: Allows users to contribute benchmark results back to the project repository without leaving the application.
  • Local-First Data Execution: Saves all performance data locally before offering the option to share, ensuring data sovereignty.
  • Customizable Testing Routines: Allows users to tweak auto-generated evaluation plans based on their specific GPU or CPU setups.
  • Direct Contribution: Allows users to seamlessly submit hardware benchmark PRs straight from the interface.

Where teams use it

Hardware Performance Evaluation

Machine learning engineers use the tool to verify if a newly purchased GPU actually delivers the expected inference speeds locally.

Optimal Model Selection

Developers run benchmarks to find the optimal balance of speed and quality among different quantizations for their specific rig.

Community Benchmark Sharing

Open-source contributors share their performance metrics to help others predict realistic model execution times on identical hardware.

Deployment Infrastructure Planning

Sysadmins test models on available hardware to determine viability before scaling up local deployment inference clusters.

Hardware Validation

Users verify new GPU clusters by confirming actual inference speeds match expected theoretical bounds.

Getting started: cargo install llmfit

README

main branch

llmfit

llmfit icon

English · 中文 · 日本語

CI Crates.io License Signed with SignPath

AlexsJones%2Fllmfit | Trendshift

Find out which open-source Large Language Models (LLMs) your hardware can comfortably run. llmfit inspects your CPU, system RAM, GPU(s), VRAM, and accelerator configuration to recommend models across popular quantizations.

📊 New: benchmark & share — real numbers from your machine, better estimates for everyone. Download a model, serve it, and measure real tok/s on your hardware — then contribute the results back to the project as a PR, straight from the TUI. No gh CLI, no third-party account. Every run is saved locally first, your own measurements replace estimates in the fit table, and each merged submission ships in the next release: anyone on identical hardware gets measured ✓ numbers before they ever run a benchmark. Follow the step-by-step benchmarking guide →

llmfit demo: searching for a model, simulating different hardware, and planning a deployment

Features

  • Hardware Auto-Detection: Detects CPU cores, system RAM, available discrete/integrated GPUs, VRAM, and unified memory architecture (NVIDIA CUDA, Apple Silicon, AMD ROCm, Intel OneAPI).
  • Model Compatibility Engine: Analyzes model parameter counts, context lengths, and quantization formats (GGUF, AWQ, GPTQ, EXL2) to project memory footprints and tokens-per-second performance.
  • Interactive TUI & Web Dashboard: Choose between a lightweight, zero-dependency terminal interface or a feature-rich web dashboard.
  • REST API Endpoint: Exposes standard HTTP JSON endpoints (/api/v1/system, /api/v1/models) for integration into orchestrators, dashboards, and automated deployment pipelines.
  • Multi-Platform Support: macOS (Apple Silicon & Intel), Linux (x86_64 & ARM64), and Windows (x86_64).
  • Hundreds of models & providers. One command to find what runs on your hardware.

A terminal tool that right-sizes LLM models to your system's RAM, CPU, and GPU. Detects your hardware, scores each model across quality, speed, fit, and context dimensions, and tells you which ones will actually run well on your machine.

Ships with an interactive TUI (default) and a classic CLI mode. Supports multi-GPU setups, MoE architectures, dynamic quantization selection, speed estimation, and local runtime providers (Ollama, llama.cpp, MLX, Docker Model Runner, LM Studio).


Sister projects

  • sympozium — managing agents in Kubernetes.
  • llmserve — a simple TUI for serving local LLM models. Pick a model, pick a backend, serve it.
  • llama-panel — a native macOS app for managing local llama-server instances.
  • llmfit-gui — a Windows desktop GUI (PowerShell + WinForms) for llmfit: browse recommendations, download into LM Studio/Ollama, and benchmark, all point-and-click.

Documentation

Get started Install · Usage · How it works
Guides TUI guide · Benchmarking step-by-step · CLI & automation · Cluster mode (Ray / multi-node) · Runtime providers · OpenClaw integration
Reference How it works (full) · Platform & GPU support · Custom models · Development
Project Contributing · Alternatives · Code signing · License

Install

Windows

WinGet
winget install --exact --id AlexsJones.llmfit
Scoop
scoop install llmfit

If Scoop is not installed, follow the Scoop installation guide.

macOS / Linux

Homebrew

Prebuilt binary (recommended, works on all macOS/Linux versions):

brew install AlexsJones/llmfit/llmfit

Or from the homebrew-core formula, which builds from source on macOS versions without a bottle:

brew install llmfit
MacPorts
port install llmfit
Quick install
curl -fsSL https://llmfit.axjns.dev/install.sh | sh

Downloads the latest release binary from GitHub and installs it to /usr/local/bin (or ~/.local/bin if no sudo).

Install to ~/.local/bin without sudo:

curl -fsSL https://llmfit.axjns.dev/install.sh | sh -s -- --local

uv / pip

To install or update llmfit:

uv tool install -U llmfit

To run without installing:

uvx llmfit

You can also install llmfit as a Python package in the normal way with tools such as pip or uv.

Pre-built Binaries

Download release binaries for Linux, macOS, and Windows directly from the GitHub Releases page. Windows binaries are only published once that release's sign-windows job has signed, repackaged, and re-checksummed them; if signing is skipped or fails, the Windows assets are held back and attached by a re-run of the release workflow (with its fresh signing requests approved) rather than shipped unsigned. Releases before v1.1.19 did not have this guard, so verify the executable signature on older versions if you require a signed binary.


Container Deployment

llmfit provides a multi-architecture Docker image (ghcr.io/alexsjones/llmfit) supporting both interactive CLI/TUI and headless Web UI / API server modes.

Interactive TUI

To launch the interactive TUI instead, pass the global --tui flag:

docker run -it --rm ghcr.io/alexsjones/llmfit --tui

Non-Interactive

This prints JSON from llmfit recommend command.

docker run ghcr.io/alexsjones/llmfit

This prints JSON from llmfit recommend command. The JSON could be further queried with jq.

podman run ghcr.io/alexsjones/llmfit recommend --use-case coding | jq '.models[].name'

To launch the interactive TUI instead, pass the global --tui flag:

docker run --rm -it ghcr.io/alexsjones/llmfit --tui

From source

git clone https://github.com/AlexsJones/llmfit.git
cd llmfit
cargo build --release
# binary is at target/release/llmfit

Usage

Terminal Interface (TUI)

Launch llmfit in your terminal without flags to start the interactive browser:

llmfit          # interactive TUI: your hardware, every model, ranked

The TUI shows your detected specs at the top and every model scored for fit, speed, quality, and context. See the TUI guide for navigation, planning, simulation, downloads, the community leaderboard, and benchmarking.

Keybindings inside the TUI:

  • b: Open community benchmarks; I: Open live inference benchmarks
  • h: Show help and keybindings
  • ↑ / ↓ or k / j: Navigate list items
  • /: Filter models by name, family, or quantization
  • Esc: Clear search / Back

Command Line Options

# Print hardware telemetry and recommended models to standard output
llmfit recommend

# Output system profile and recommendations in raw JSON format
llmfit recommend --json

# Estimate SSD capacity for keeping three runnable models
llmfit storage --keep 3 --selection largest --json

# Start the native HTTP API server
llmfit serve --host 0.0.0.0 --port 8787

See model library storage for selection, OS reserve, download scratch, free-space headroom, and hardware simulation.

Web UI & API Server

docker run -d -p 8787:8787 ghcr.io/alexsjones/llmfit serve
Docker Compose
---
services:
  llmfit:
    image: ghcr.io/alexsjones/llmfit:latest
    container_name: llmfit
    restart: unless-stopped
    command: ["serve", "--host", "0.0.0.0", "--port", "8787"]
    ports:
      - "8787:8787"
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8787/health"]
      interval: 15s
      timeout: 5s
      retries: 3
      start_period: 10s

For scripts, agents, and classic terminal output:

llmfit fit                    # table of all models ranked by fit
llmfit fit --show-incompatible # also show backend-incompatible models as non-runnable
llmfit recommend --json       # top picks as JSON (agent/script consumption)
llmfit info "<model>"         # one model: fit analysis, estimate basis, verify commands
llmfit bench                  # measure real tok/s/TTFT against your running provider
llmfit doctor                 # hardware detection report for bug reports
llmfit serve                  # start the api and web user interface

Full reference: CLI & automation.

Cluster mode (Ray / multi-node)

Preview: so far tested only against a mock Ray Dashboard. If you run Ray, please share the output of curl <dashboard>/nodes?view=summary in an issue.

Check which models fit across several machines, such as 3 × DGX Spark or a couple of multi-GPU servers running Ray and vLLM:

  • Know before you launch: find out whether a model fits across your machines before you spend an hour starting vLLM.
  • Catch splits vLLM would refuse: most models can't be split 3 ways across 3 GPUs. llmfit picks a split vLLM accepts, or says none fits, instead of letting vLLM crash at startup.
  • One reviewable file: Ray discovery writes a single cluster.toml, so you don't have to tally VRAM by hand.
  • Changes nothing: llmfit reads Ray's dashboard once and never launches or schedules anything.
llmfit cluster init --ray-url http://<head>:8265   # or hand-write cluster.toml
llmfit cluster status
llmfit --cluster fit

See the cluster guide for walkthroughs, Ray token handling and today's limits.


Community & Benchmarks

llmfit includes hardware detection and performance benchmarks contributed by the community. You can share your hardware benchmark results using:

llmfit bench --share

How it works

llmfit detects your hardware (RAM, CPU, GPU/VRAM, backend), then scores every model in its catalog across four dimensions: memory fit, estimated speed, quality, and context. Speed estimates come from a memory-bandwidth model grounded in runtime sampling and real community measurements — and every estimate ships its inputs, so llmfit info shows exactly what a number assumes and how to verify it on your machine.

Full detail, including the estimation formulas and the model database: How llmfit works.


Contributing

Contributions are welcome, especially new models.

Before submitting a PR

Please run cargo fmt before pushing your changes. Most CI check failures are caused by unformatted code:

cargo fmt

Guides for adding models — locally (no rebuild) or to the built-in catalog: Custom models.


Alternatives

If you're looking for a different approach, check out llm-checker -- a Node.js CLI tool with Ollama integration that can pull and benchmark models directly. It takes a more hands-on approach by actually running models on your hardware via Ollama, rather than estimating from specs. Good if you already have Ollama installed and want to test real-world performance. Note that it doesn't support MoE (Mixture-of-Experts) architectures -- all models are treated as dense, so memory estimates for models like Mixtral or DeepSeek-V3 will reflect total parameter count rather than the smaller active subset.


Code signing

llmfit's Windows release binaries are intended to be digitally signed (Authenticode) via SignPath.io, with a free code signing certificate provided by the SignPath Foundation. The release workflow only attaches Windows assets after the sign-windows job has signed, repackaged, and re-checksummed them; when signing is skipped or fails, the Windows zips (and the PyPI Windows wheels built from them) are held back until the workflow is re-run against the tag and the new signing requests that re-run submits are approved. Releases before v1.1.19 could publish an unsigned Windows artifact, so verify the executable signature on those before relying on it.

Signing happens automatically in the release pipeline: only artifacts built by GitHub Actions from this repository are submitted for signing, and signing requests are approved by the project maintainer (@AlexsJones).

Code signing policy: see the SignPath Foundation code signing policy and terms.

Privacy: this program will not transfer any information to other networked systems unless specifically requested by the user or the person installing or operating it. llmfit only contacts external services when you explicitly use the corresponding feature (e.g. model downloads, runtime provider queries, or the community leaderboard).


License

MIT

View on GitHub

Recent activity

commits and pull requests

Releases and announcements

138 total
  1. v1.1.20v1.1.20Nov 3, 202610.6K downloads

    ## [1.1.20](https://github.com/AlexsJones/llmfit/compare/v1.1.19...v1.1.20) (2026-11-03) ### Bug Fixes * NVFP4 and MXFP4 safetensors repos are no longer offered as llama.cpp fits, so they don't top the list on ROCm and CPU machines ([#1230](https://github.com/AlexsJones/llmfit/issues/1230)) * 39 large models now have real attention-head counts from their `config.json` (including Llama-3.1-405B and MiniMax-M1), so `--cluster` can check their tensor-parallel splits ([#1230](https://github.com/AlexsJones/llmfit/issues/1230)) * `diff` and the TUI compare view no longer put a model that doesn't fit, or whose cluster split is refused, ahead of one that fits ([#1230](https://github.com/AlexsJones/llmfit/issues/1230)) * **cluster:** the banner shows the VRAM pool actually used when `--memory` is set ([#1230](https://github.com/AlexsJones/llmfit/issues/1230)) * **cluster:** Ray discovery sets `backend` from the GPU vendor where Ray reports it, and otherwise names the nodes that need `backend` set instead of assuming CUDA ([#1230](https://github.com/AlexsJones/llmfit/issues/1230)) * **cluster:** `--json` errors from `cluster init`, `cluster status`, `cluster clear` and `--cluster`; "1 nod

  2. v1.1.19v1.1.19Nov 3, 2026659 downloads

    ## [1.1.19](https://github.com/AlexsJones/llmfit/compare/v1.1.18...v1.1.19) (2026-11-03) ### Features * cluster mode (preview): find models that fit a Ray or multi-node GPU cluster — see [docs/cluster.md](https://github.com/AlexsJones/llmfit/blob/main/docs/cluster.md) ([#1224](https://github.com/AlexsJones/llmfit/issues/1224), [#1225](https://github.com/AlexsJones/llmfit/issues/1225), [#1228](https://github.com/AlexsJones/llmfit/issues/1228); takes over [#279](https://github.com/AlexsJones/llmfit/issues/279) by [@michaeljabbour](https://github.com/michaeljabbour)) * distinguish exact and legacy PCIe benchmark matches ([#1211](https://github.com/AlexsJones/llmfit/issues/1211)) ([52a6f49](https://github.com/AlexsJones/llmfit/commit/52a6f49c3851ab0146d2a78277bd233af87399b8)) ### Bug Fixes * exclude NVFP4 on ROCm GPUs ([#1203](https://github.com/AlexsJones/llmfit/issues/1203)) ([e3d5fe5](https://github.com/AlexsJones/llmfit/commit/e3d5fe56ae35bb344dca8ee348b935f9ea294da0)) * **hardware:** avoid collecting unrelated process information ([#1200](https://github.com/AlexsJones/llmfit/issues/1200)) ([a560b35](https://github.com/AlexsJones/llmfit/commit/a560b353a1fd1132fabe3b9da2844a54

  3. v1.1.18v1.1.18Oct 28, 20267.1K downloads

    ## [1.1.18](https://github.com/AlexsJones/llmfit/compare/v1.1.17...v1.1.18) (2026-10-28) ### Features * **bench:** discover and benchmark models loaded in LM Studio ([#1129](https://github.com/AlexsJones/llmfit/issues/1129)) ([ccc1390](https://github.com/AlexsJones/llmfit/commit/ccc1390757b23398b3e81964debeef75e41a2be4)) * **cli:** show backend-incompatible models on request ([#1143](https://github.com/AlexsJones/llmfit/issues/1143)) ([eb8db38](https://github.com/AlexsJones/llmfit/commit/eb8db3835157d1c300ffa9050ac58f63df11617c)) * **hardware:** detect and show the PCIe link of NVIDIA GPUs ([#1153](https://github.com/AlexsJones/llmfit/issues/1153)) ([b8124fb](https://github.com/AlexsJones/llmfit/commit/b8124fb6c4ae9a943dac86440020dc0bdefdada1)) * **scraper:** backfill release_date for catalog entries without one ([#1060](https://github.com/AlexsJones/llmfit/issues/1060)) ([31a90a4](https://github.com/AlexsJones/llmfit/commit/31a90a4465f2f5658589bdfd2eedd2e77aaa28d4)), closes [#176](https://github.com/AlexsJones/llmfit/issues/176) * **web:** catch the dashboard up with estimate confidence, concurrency, storage and TUI filters ([#1094](https://github.com/AlexsJones/llmfit/issues/1

  4. v1.1.17v1.1.17Oct 8, 202622K downloads

    ## [1.1.17](https://github.com/AlexsJones/llmfit/compare/v1.1.16...v1.1.17) (2026-10-08) ### Features * **cli:** add --memory-percent and --ram-percent overrides ([#1089](https://github.com/AlexsJones/llmfit/issues/1089)) ([17d45bd](https://github.com/AlexsJones/llmfit/commit/17d45bd667bb9cca95008182e07852d2208087c1)) * **tui:** add mouse support to existing controls ([#1110](https://github.com/AlexsJones/llmfit/issues/1110)) ([8caf925](https://github.com/AlexsJones/llmfit/commit/8caf9256739d8d3e4f5359462990f8cf18f1856c)) ### Bug Fixes * **api:** make include_too_tight=true return too_tight rows ([#1087](https://github.com/AlexsJones/llmfit/issues/1087)) ([cbd30d7](https://github.com/AlexsJones/llmfit/commit/cbd30d7d512885c1eff2e545da37c09f2768963c)) * **cli:** honor --json in search ([#1113](https://github.com/AlexsJones/llmfit/issues/1113)) ([f89395b](https://github.com/AlexsJones/llmfit/commit/f89395bed72a3d4905e792688d8119141f3b792e)), closes [#901](https://github.com/AlexsJones/llmfit/issues/901) * **cli:** reject unknown recommend filter values ([#1114](https://github.com/AlexsJones/llmfit/issues/1114)) ([35fdd4f](https://github.com/AlexsJones/llmfit/commit/35fdd4f941f7

  5. v1.1.16v1.1.16Sep 19, 202623.6K downloads

    ## [1.1.16](https://github.com/AlexsJones/llmfit/compare/v1.1.15...v1.1.16) (2026-09-19) ### Features * **concurrency:** concurrent-session capacity estimator ([#140](https://github.com/AlexsJones/llmfit/issues/140)) ([#999](https://github.com/AlexsJones/llmfit/issues/999)) ([1936acc](https://github.com/AlexsJones/llmfit/commit/1936acc7fbd2fad0a5e7a2fe7510f4004d4007e4)) * **models:** Sept 2026 model refresh (GLM-5.3, Qwen3.8-Flash-Next, DeepSeek-V4.1, Kimi-K3 and more) ([#1055](https://github.com/AlexsJones/llmfit/issues/1055)) ([6dd1f3c](https://github.com/AlexsJones/llmfit/commit/6dd1f3cd3098cc0e2abe9e4a1ccdaa518e614aeb)) * recognize native ternary (1.58-bit) models ([#886](https://github.com/AlexsJones/llmfit/issues/886)) ([569a9ac](https://github.com/AlexsJones/llmfit/commit/569a9ac6cf80679631fb4a12830f1b832c4f5f30)) * **storage:** add disk planning for model libraries ([#1023](https://github.com/AlexsJones/llmfit/issues/1023)) ([42c2641](https://github.com/AlexsJones/llmfit/commit/42c2641b75ba610997718b90231ef2a412b66bf3)) ### Bug Fixes * **bench:** error when the requested model is not available ([#1041](https://github.com/AlexsJones/llmfit/issues/1041)) ([f66aad9](http

Code frequency

additions and deletions
+298K-298KWeek of 2026-02-15: +43,404 linesWeek of 2026-02-15: -23,436 linesWeek of 2026-02-22: +31,795 linesWeek of 2026-02-22: -12,060 linesWeek of 2026-03-01: +46,039 linesWeek of 2026-03-01: -34,334 linesWeek of 2026-03-08: +5,759 linesWeek of 2026-03-08: -935 linesWeek of 2026-03-15: +18,812 linesWeek of 2026-03-15: -6,446 linesWeek of 2026-03-22: +1,155 linesWeek of 2026-03-22: -371 linesWeek of 2026-03-29: +96,484 linesWeek of 2026-03-29: -76,299 linesWeek of 2026-04-05: +7,817 linesWeek of 2026-04-05: -4,360 linesWeek of 2026-04-12: +2,994 linesWeek of 2026-04-12: -617 linesWeek of 2026-04-19: +22,161 linesWeek of 2026-04-19: -55,397 linesWeek of 2026-04-26: +64,643 linesWeek of 2026-04-26: -6,099 linesWeek of 2026-05-03: +45,567 linesWeek of 2026-05-03: -17,542 linesWeek of 2026-05-10: +64,846 linesWeek of 2026-05-10: -20,949 linesWeek of 2026-05-17: +284,674 linesWeek of 2026-05-17: -74,965 linesWeek of 2026-05-24: +1,057 linesWeek of 2026-05-24: -516 linesWeek of 2026-05-31: +1,152 linesWeek of 2026-05-31: -138 linesWeek of 2026-06-07: +40,023 linesWeek of 2026-06-07: -9,148 linesWeek of 2026-06-14: +32,678 linesWeek of 2026-06-14: -20,953 linesWeek of 2026-06-21: +42,066 linesWeek of 2026-06-21: -17,506 linesWeek of 2026-06-28: +14,786 linesWeek of 2026-06-28: -200,645 linesWeek of 2026-07-05: +26,144 linesWeek of 2026-07-05: -14,118 linesWeek of 2026-07-12: +808 linesWeek of 2026-07-12: -200 linesWeek of 2026-07-19: +1,851 linesWeek of 2026-07-19: -137 linesWeek of 2026-07-26: +1,605 linesWeek of 2026-07-26: -57 linesWeek of 2026-08-02: +298,004 linesWeek of 2026-08-02: -124,365 linesWeek of 2026-08-09: +70,293 linesWeek of 2026-08-09: -29,569 linesWeek of 2026-08-16: +80,556 linesWeek of 2026-08-16: -40,247 linesWeek of 2026-08-23: +123,046 linesWeek of 2026-08-23: -77,282 linesWeek of 2026-08-30: +11,842 linesWeek of 2026-08-30: -3,845 linesWeek of 2026-09-06: +6,865 linesWeek of 2026-09-06: -196 linesWeek of 2026-09-13: +278,188 linesWeek of 2026-09-13: -183,466 linesWeek of 2026-09-20: +345 linesWeek of 2026-09-20: -5 linesWeek of 2026-09-27: +72,776 linesWeek of 2026-09-27: -37,951 linesWeek of 2026-10-04: +73,064 linesWeek of 2026-10-04: -34,563 linesWeek of 2026-10-11: +120,126 linesWeek of 2026-10-11: -47,696 linesWeek of 2026-10-18: +886 linesWeek of 2026-10-18: -89 linesWeek of 2026-10-25: +99,280 linesWeek of 2026-10-25: -43,675 linesWeek of 2026-11-01: +4,698 linesWeek of 2026-11-01: -965 linesWeek of 2026-11-08: +1,135 linesWeek of 2026-11-08: -233 linesFeb 15, 2026Nov 8, 2026
+2.1M lines added, -1.2M removed over the last year.

Commits per week

last 52 weeks
950Week of 2025-11-16: 0 commitsWeek of 2025-11-23: 0 commitsWeek of 2025-11-30: 0 commitsWeek of 2025-12-07: 0 commitsWeek of 2025-12-14: 0 commitsWeek of 2025-12-21: 0 commitsWeek of 2025-12-28: 0 commitsWeek of 2026-01-04: 0 commitsWeek of 2026-01-11: 0 commitsWeek of 2026-01-18: 0 commitsWeek of 2026-01-25: 0 commitsWeek of 2026-02-01: 0 commitsWeek of 2026-02-08: 0 commitsWeek of 2026-02-15: 95 commitsWeek of 2026-02-22: 69 commitsWeek of 2026-03-01: 70 commitsWeek of 2026-03-08: 36 commitsWeek of 2026-03-15: 48 commitsWeek of 2026-03-22: 18 commitsWeek of 2026-03-29: 25 commitsWeek of 2026-04-05: 23 commitsWeek of 2026-04-12: 23 commitsWeek of 2026-04-19: 38 commitsWeek of 2026-04-26: 23 commitsWeek of 2026-05-03: 14 commitsWeek of 2026-05-10: 8 commitsWeek of 2026-05-17: 21 commitsWeek of 2026-05-24: 12 commitsWeek of 2026-05-31: 15 commitsWeek of 2026-06-07: 27 commitsWeek of 2026-06-14: 13 commitsWeek of 2026-06-21: 19 commitsWeek of 2026-06-28: 41 commitsWeek of 2026-07-05: 60 commitsWeek of 2026-07-12: 28 commitsWeek of 2026-07-19: 28 commitsWeek of 2026-07-26: 5 commitsWeek of 2026-08-02: 25 commitsWeek of 2026-08-09: 29 commitsWeek of 2026-08-16: 28 commitsWeek of 2026-08-23: 32 commitsWeek of 2026-08-30: 29 commitsWeek of 2026-09-06: 13 commitsWeek of 2026-09-13: 35 commitsWeek of 2026-09-20: 5 commitsWeek of 2026-09-27: 9 commitsWeek of 2026-10-04: 26 commitsWeek of 2026-10-11: 22 commitsWeek of 2026-10-18: 10 commitsWeek of 2026-10-25: 52 commitsWeek of 2026-11-01: 31 commitsWeek of 2026-11-08: 8 commitsNov 16, 2025Nov 8, 2026
1.1K commits in the last 52 weeks.

When work happens

weekday and hour
SunMonTueWedThuFriSat036912151821Sun 0:00 — 2 commitsSun 1:00 — 2 commitsSun 2:00 — 1 commitsSun 3:00 — 2 commitsSun 4:00 — 1 commitsSun 5:00 — 1 commitsSun 6:00 — 1 commitsSun 7:00 — 8 commitsSun 8:00 — 8 commitsSun 9:00 — 6 commitsSun 10:00 — 6 commitsSun 11:00 — 1 commitsSun 12:00 — 6 commitsSun 13:00 — 10 commitsSun 14:00 — 6 commitsSun 15:00 — 20 commitsSun 16:00 — 19 commitsSun 17:00 — 9 commitsSun 18:00 — 7 commitsSun 19:00 — 3 commitsSun 20:00 — 5 commitsSun 21:00 — 7 commitsSun 22:00 — 4 commitsSun 23:00 — 3 commitsMon 0:00 — 36 commitsMon 1:00 — 6 commitsMon 2:00 — 4 commitsMon 3:00 — 4 commitsMon 4:00 — 0 commitsMon 5:00 — 1 commitsMon 6:00 — 8 commitsMon 7:00 — 9 commitsMon 8:00 — 26 commitsMon 9:00 — 5 commitsMon 10:00 — 9 commitsMon 11:00 — 18 commitsMon 12:00 — 8 commitsMon 13:00 — 17 commitsMon 14:00 — 16 commitsMon 15:00 — 19 commitsMon 16:00 — 4 commitsMon 17:00 — 8 commitsMon 18:00 — 8 commitsMon 19:00 — 4 commitsMon 20:00 — 11 commitsMon 21:00 — 7 commitsMon 22:00 — 5 commitsMon 23:00 — 1 commitsTue 0:00 — 1 commitsTue 1:00 — 1 commitsTue 2:00 — 0 commitsTue 3:00 — 2 commitsTue 4:00 — 0 commitsTue 5:00 — 0 commitsTue 6:00 — 5 commitsTue 7:00 — 10 commitsTue 8:00 — 9 commitsTue 9:00 — 8 commitsTue 10:00 — 10 commitsTue 11:00 — 35 commitsTue 12:00 — 20 commitsTue 13:00 — 12 commitsTue 14:00 — 18 commitsTue 15:00 — 10 commitsTue 16:00 — 4 commitsTue 17:00 — 6 commitsTue 18:00 — 4 commitsTue 19:00 — 1 commitsTue 20:00 — 7 commitsTue 21:00 — 9 commitsTue 22:00 — 5 commitsTue 23:00 — 2 commitsWed 0:00 — 2 commitsWed 1:00 — 3 commitsWed 2:00 — 3 commitsWed 3:00 — 4 commitsWed 4:00 — 4 commitsWed 5:00 — 1 commitsWed 6:00 — 8 commitsWed 7:00 — 10 commitsWed 8:00 — 19 commitsWed 9:00 — 12 commitsWed 10:00 — 5 commitsWed 11:00 — 8 commitsWed 12:00 — 5 commitsWed 13:00 — 3 commitsWed 14:00 — 8 commitsWed 15:00 — 10 commitsWed 16:00 — 6 commitsWed 17:00 — 5 commitsWed 18:00 — 13 commitsWed 19:00 — 4 commitsWed 20:00 — 5 commitsWed 21:00 — 7 commitsWed 22:00 — 6 commitsWed 23:00 — 6 commitsThu 0:00 — 2 commitsThu 1:00 — 3 commitsThu 2:00 — 1 commitsThu 3:00 — 1 commitsThu 4:00 — 1 commitsThu 5:00 — 1 commitsThu 6:00 — 3 commitsThu 7:00 — 10 commitsThu 8:00 — 9 commitsThu 9:00 — 18 commitsThu 10:00 — 15 commitsThu 11:00 — 7 commitsThu 12:00 — 10 commitsThu 13:00 — 2 commitsThu 14:00 — 10 commitsThu 15:00 — 4 commitsThu 16:00 — 4 commitsThu 17:00 — 11 commitsThu 18:00 — 4 commitsThu 19:00 — 5 commitsThu 20:00 — 5 commitsThu 21:00 — 13 commitsThu 22:00 — 7 commitsThu 23:00 — 2 commitsFri 0:00 — 4 commitsFri 1:00 — 1 commitsFri 2:00 — 5 commitsFri 3:00 — 3 commitsFri 4:00 — 3 commitsFri 5:00 — 8 commitsFri 6:00 — 8 commitsFri 7:00 — 16 commitsFri 8:00 — 14 commitsFri 9:00 — 11 commitsFri 10:00 — 13 commitsFri 11:00 — 20 commitsFri 12:00 — 4 commitsFri 13:00 — 2 commitsFri 14:00 — 9 commitsFri 15:00 — 5 commitsFri 16:00 — 4 commitsFri 17:00 — 4 commitsFri 18:00 — 5 commitsFri 19:00 — 1 commitsFri 20:00 — 1 commitsFri 21:00 — 4 commitsFri 22:00 — 3 commitsFri 23:00 — 3 commitsSat 0:00 — 2 commitsSat 1:00 — 1 commitsSat 2:00 — 2 commitsSat 3:00 — 5 commitsSat 4:00 — 0 commitsSat 5:00 — 2 commitsSat 6:00 — 1 commitsSat 7:00 — 2 commitsSat 8:00 — 7 commitsSat 9:00 — 13 commitsSat 10:00 — 7 commitsSat 11:00 — 16 commitsSat 12:00 — 6 commitsSat 13:00 — 6 commitsSat 14:00 — 4 commitsSat 15:00 — 7 commitsSat 16:00 — 4 commitsSat 17:00 — 10 commitsSat 18:00 — 1 commitsSat 19:00 — 1 commitsSat 20:00 — 6 commitsSat 21:00 — 7 commitsSat 22:00 — 0 commitsSat 23:00 — 1 commits
Commit volume by weekday and hour (UTC). Larger dots mean more commits.

Who is committing

last 52 weeks
Maintainer commits698 (51%)
Community commits681 (49%)

1,379 commits in total over the last year.

DateListRankStars gained
Mar 5, 2026daily#23+136
Mar 3, 2026daily#15+220
Mar 2, 2026daily#8+395
Mar 1, 2026daily#22+209
Feb 27, 2026daily#22+157
Feb 26, 2026daily#22+172
Feb 25, 2026daily#17+192
Feb 24, 2026daily#25+125
Feb 23, 2026daily#9+235
Feb 22, 2026daily#7+369
Feb 21, 2026daily#17+171
  • affaan-m/ECC

    The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

    307.9K stars · JavaScript

  • NousResearch/hermes-agent

    The agent that grows with you

    265K stars · Python

  • firecrawl/firecrawl

    Knowledge infrastructure to make your agents smarter. Search, scrape and access 550+ data providers through one API.

    205.3K stars · TypeScript

  • DietrichGebert/ponytail

    Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.

    195.7K stars · JavaScript

  • ultraworkers/claw-code

    An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.

    194.8K stars · Rust

  • Significant-Gravitas/AutoGPT

    AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.

    188.1K stars · Python