# beclab/Olares

Open-Source Personal Cloud OS for Always-On Agents

Repository: https://github.com/beclab/Olares
Canonical: https://ross.abutalabs.com/products/olares
Homepage: https://www.olares.com/
Language: Go
License: AGPL-3.0
License Family: copyleft
Topics: kubernetes, self-hosted, homelab, edge-ai, local-ai, ai-agents, model-serving, personal-cloud, agentic-ai, cloud-native, data-privacy, operating-system
Last push: 2026-08-26T19:40:26+00:00

## Health v2 (maintenance only)
Score: 89/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 93, longevity 61
- inputs: {"age_days": 856, "days_push": 7, "days_rel": 48, "gap_med": 14.5, "n_releases_24m": 25}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 5242, forks 321 (observed 2026-08-28T04:09:13.559376+00:00)

## What it is
Olares is an open-source personal cloud operating system built on Kubernetes that turns your own hardware into a self-hosted AI platform for running AI agents and local LLMs. It provides a browser-accessible desktop experience with one-click app installs, GPU/accelerator pooling, file storage, identity, VPN access, and natural-language management via Agent Skills and the olares-cli.

## Use cases
- run local LLMs and AI agents on my own hardware
- self-host a personal cloud with files, apps, and AI in one place
- manage my homelab server in plain language with an AI agent
- pool GPUs across nodes for AI and media workloads
- host open-source apps with one-click installs from an app market
- keep my AI data private instead of using cloud AI services
- give my family or small team a private cloud with shared apps

## When to choose
- you want to run AI agents and local models on hardware you own with full data privacy
- you want a Kubernetes-backed homelab OS with a polished desktop, app market, and storage
- you want to operate your server through natural-language agent commands
- you need GPU sharing features like time-slicing and memory-slicing for AI workloads

## When to avoid
- you need a lightweight single-purpose tool rather than a full OS-level platform
- you cannot dedicate always-on hardware or lack resources for a Kubernetes-based stack
- you require strict enterprise compliance or a permissive license, since Olares is AGPL-3.0
- you only want managed cloud hosting with zero self-hosting responsibilities

## Facets
- artifact type: application
- maturity: active
- function: self-hosted, llm-inference, agent-framework, container-orchestration, file-system, networking, security, cli, gpu-computing, chatbot
- domain: self-hosted, artificial-intelligence, large-language-models, cloud-computing, privacy, developer-tools, operating-systems
- platform: self-hosted, cli
- tags: personal-cloud-os, homelab, local-ai, edge-ai, data-ownership, app-market, agentic-ai, model-serving, natural-language-management, agent-skills, ai-agents, containers, linux, kubernetes, docker, gpu, web-server

## Member repositories
- beclab/Olares (main) score 89

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:13.559376+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 2026-08-29T17:59:36.493687+00:00, confidence not recorded.
  - readme: https://github.com/beclab/Olares (fetched 2026-08-28T04:09:13.559376+00:00, sha f1685771eecb)
  - homepage: https://www.olares.com/ (fetched 2026-08-29T08:54:46.121713+00:00, sha ff33bfeb84ce)
  - site_page: https://www.olares.com/docs/manual/get-started (fetched 2026-08-29T08:54:46.130869+00:00, sha 1d118ad150db)
  - site_page: https://www.olares.com/docs/manual/overview (fetched 2026-08-29T08:54:46.132505+00:00, sha d2a88e501ba8)
  - site_page: https://www.olares.com/about-us (fetched 2026-08-29T08:54:46.137873+00:00, sha 76550db62ad8)
  - site_page: https://www.olares.com/changelog (fetched 2026-08-29T08:54:46.134515+00:00, sha db3d7422113b)
- Data as of 2026-08-30T08:39:29.467469+00:00.