#5555th of 71 in Model serving
olares
Kubernetes-based personal cloud OS for running local LLMs and AI agents
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 5.4k
- License
- AGPL-3.0
- Last commit
- Nov 2026
- Last release
- Sep 2026
- Language
- Go
4,286 on 21 Apr 20265,357 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Olares installs a Kubernetes-based personal cloud on your own machines and manages apps, files, networking and GPUs from a browser desktop. It ships an app market with local AI apps and models, a Model Console for local LLMs, and the Lares assistant, and it can host agents such as OpenClaw. The olares-cli tool and Agent Skills let AI agents install and operate the system.
Who it is for: Self-hosters wanting a home or small-team cloud with local LLMs and agents
Strengths
- Pools GPUs across nodes with time-slicing, memory-slicing and exclusive modes
- Bundles files, VPN, reverse proxy, Vault and app market in one system
- Multiple install paths: shell script, ISO image, Docker Compose
- olares-cli with Agent Skills lets agents install and operate the system
Weaknesses
- Needs 4 CPU cores, 8 GB RAM and 150 GB SSD; installation fails on HDD
- Script install supports only Ubuntu 22.04 to 25.04 and Debian 12 or 13
- Full Kubernetes stack is heavy for someone who only wants to serve one model
- Installer is run as curl piped to bash with sudo
What it needs
- RAM ≥ 8 GB
- GPU optional
- Needs Ubuntu or Debian Linux, Kubernetes (bundled), 150 GB SSD
- Models: local LLMs via Model Console and Olares Market
Also in Model serving
See all 71| Rank | Project | Score |
|---|---|---|
| 1 | llama.cppC/C++ inference engine for GGUF LLMs and VLMs with an OpenAI-compatible server | 94 out of 100 |
| 2 | OllamaLocal model runner with CLI and REST API for open-weight LLMs | 92 out of 100 |
| 3 | LocalAISelf-hosted inference server with OpenAI-compatible API and pluggable backends | 91 out of 100 |
| #4 | vLLMInference engine serving LLMs with PagedAttention and an OpenAI-compatible API | 88 out of 100 |
| #5 | strataRuns a 125B-parameter MoE model on one consumer GPU plus system RAM | 88 out of 100 |