Enterprise DNAEnterprise DNA

Open Source

OrcaRouter Lite

by Continuum AI

A self-hosted, single-workspace model router that accepts OpenAI, Anthropic and Gemini style requests and, in auto mode, picks the cheapest configured model that can handle each request.

Our verdict

Source reviewedReviewed 22 Oct 2026 by Enterprise DNA editorial

A self-hosted, single-workspace model router that accepts OpenAI, Anthropic and Gemini style requests and, in auto mode, picks the cheapest configured model that can handle each request.

Best for: Developers who want a small router with a local dashboard in front of their own provider keys.

Choose something else if: you need teams, single sign-on, audit logs or more than one server. Lite leaves those out, and its only tagged release dates from May 2026.

Source review

Using OrcaRouter Lite in practice

Where it runs

Runs on your own server, with SQLite by default. If you add a hosted OrcaRouter key, some requests go to Continuum AI's service.

How to install

Needs git, Docker Compose and one provider key. Clone the repository, copy .env.example to .env, add the key and run docker compose up. Point clients at http://localhost:8000/v1 with the sk-orca key printed at startup.

Install commands

git clone https://github.com/Continuum-AI-Corp/OrcaRouter-Lite.git
cd OrcaRouter-Lite
cp .env.example .env
docker compose up

As published in the linked sources. Not run by us.

What it costs

Lite is free under the MIT licence, and you pay your providers and your hosting. The optional hosted OrcaRouter bills per token: its Hacker plan is free with no token markup, Team is US$499 a month and Enterprise is quoted.

Prices and plans change. Check the linked sources before committing.

Why consider it

  • OpenAI-compatible, Anthropic and Gemini endpoints, with a cheapest-capable auto mode.
  • Local dashboard and analytics, with no telemetry leaving the server according to its README.

What to watch for

  • The quickstart builds from the main branch because no image is published yet.
  • Set CREDENTIAL_ENCRYPTION_KEY before saving provider keys in the dashboard. The example file leaves it commented out.
  • It installs the LiteLLM library without pinning a version, so check which version a build pulled in.

A first task to try

Add one provider key, send a request with the model set to auto, then check in the dashboard which model it chose and what it cost.

Suggested exercise, not an installation test. We reviewed the linked documentation; no host compatibility is implied.

Other options to understand

Related tools can solve different parts of the same job. Compare fit before treating them as replacements.

Key facts

What we have checked

Latest version
0.1.0 (released 11 May 2026)
Checked GitHub
Days since the last commit
29
Checked GitHub
GitHub forks
292
Checked GitHub
GitHub stars
1,648
Checked GitHub
Days to a first reply on issues
42.2
Checked GitHub
Open issues
35
Checked GitHub
Releases in the last 90 days
0
Checked GitHub

Each value shows when we last confirmed it and where it came from. Values we have not confirmed are left out, never guessed.

Stars and usage

How OrcaRouter Lite is growing

GitHub stars

1,714 to 1,648

stars. 1 Oct 2026 to 6 Nov 2026, 3 readings.

Source: GitHub. Checked .

Health and security

Is OrcaRouter Lite looked after, and what do its setup files show

Health score

56 out of 100

Does someone look after this repository? It does not say the code is good or safe. Worked out from GitHub signals checked . How it is worked out

PartPointsWhat we saw
Recent activity30 of 30Last push 29 days ago
Releases0 of 150 releases in 90 days
Issue replies0 of 20First reply in 42.2 days (median)
Maintainer15 of 15Owned by an organisation
Official sourceNot countedMCP servers only

Sources

Where this comes from

A source shows what was published when we checked it. It does not certify security, performance or a working installation. Read our method.

Changelog

What changed

  1. Released 0.1.0

    Source: GitHub

Free 27-page guide

Get the free Developer’s Field Guide

A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.

Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.

No spam. Unsubscribe any time.