# magnitudedev/magnitude

Open source agent with local models built in. Fully private and offline. Works out of the box on any hardware.

Repository: https://github.com/magnitudedev/magnitude
Canonical: https://ross.abutalabs.com/products/magnitudedev-magnitude
Homepage: https://magnitude.dev
Language: TypeScript
License: Apache-2.0
License Family: permissive
Last push: 2026-08-26T19:20:03+00:00

## Health v2 (maintenance only)
Score: 80/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 98, longevity 5
- inputs: {"age_days": 82, "days_push": 7, "days_rel": 12, "gap_med": 1.0, "n_releases_24m": 7}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1510, forks 113 (observed 2026-08-28T04:04:55.731652+00:00)

## What it is
Magnitude is an open-source CLI tool and agent that bundles local LLM inference with an agent harness, profiling your hardware to recommend, download, and tune the best models automatically. Everything runs fully offline and privately on your machine with no API keys or token costs.

## Use cases
- run an AI coding agent fully offline with local models
- automatically pick the best local LLM for my hardware
- replace Ollama with a zero-config local inference setup
- keep AI agent prompts and files private on my machine
- run local models without API keys or token costs
- extend an agent with skills for Excel, PDFs, and Chrome
- serve local models to harnesses like Claude Code or Cline

## When to choose
- you want a private, offline AI agent with no inference setup
- you're unsure which local model or quantization fits your machine
- you want an inference server tuned for agent workloads with just-in-time model loading
- you want to avoid token costs, rate limits, and subscriptions

## When to avoid
- you need frontier model quality only available via cloud APIs
- you're on Windows without WSL
- your hardware has too little memory for useful local models
- you want fine-grained manual control over model servers and quantization choices

## Facets
- artifact type: cli-tool
- maturity: active
- function: llm-inference, agent-framework, cli, chatbot
- domain: artificial-intelligence, large-language-models, developer-tools, privacy
- platform: cli, self-hosted
- tags: local-llm, offline-ai, hardware-profiling, model-management, inference-server, agent-harness, skills, command-line, macos, linux, nodejs

## Member repositories
- magnitudedev/magnitude (main) score 80

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:55.731652+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:32:26.703512+00:00, confidence not recorded.
  - readme: https://github.com/magnitudedev/magnitude (fetched 2026-08-28T04:04:55.731652+00:00, sha d0aee71bb28e)
  - homepage: https://magnitude.dev (fetched 2026-08-29T11:36:43.040482+00:00, sha 505d930a30a1)
  - site_page: https://docs.magnitude.dev (fetched 2026-08-29T11:36:43.049810+00:00, sha d1399058897e)
  - site_page: https://docs.magnitude.dev/models (fetched 2026-08-29T11:36:43.051818+00:00, sha 2683caae26b7)
- Data as of 2026-08-30T08:39:29.467469+00:00.