Integration guide

Connect Roo Code to OpenLux

Roo Code supports an OpenAI-compatible provider directly in its settings. Enter the OpenLux Base URL and key, add a catalogue model ID, then use Roo Code with native tool calling across GPT, Claude, Gemini and DeepSeek models.

Create an API key Read the docs

Abstract orange glass sculpture illustrating the Roo Code integration with OpenLux — Roo Code + OpenLux — one key, one endpoint, one bill.

Set it up in four steps

Install the Roo Code extension

Install Roo Code in VS Code and open the extension settings using the gear icon.

Create an OpenLux key

Create a key in the OpenLux console. Use a separate key for Roo Code so its usage can be identified in the request log.

Configure OpenAI Compatible

Choose OpenAI Compatible as the API provider, enter https://openlux.lol/v1 as the Base URL, paste the key and add a model ID from the OpenLux catalogue.

Verify native tool calling

Run a small task that needs file reading, editing and a command. Roo Code requires native tool calling, so choose a model tagged Tools if the task does not execute as expected.

Configuration

Copy these blocks, replace the placeholder key and you are done.

Roo Code provider settings

API Provider:  OpenAI Compatible
Base URL:      https://openlux.lol/v1
API Key:       sk-openlux-...
Model:         gpt-6-astra

# other valid model IDs
claude-sonnet-4-5-20250929
gemini-3.8-flash
deepseek-v4

Models that fit Roo Code

GPT-6 Astra

OpenAI · $0.3677 / 1M tokens

Claude Sonnet 4.5

Anthropic · $0.2647 / 1M tokens

Gemini 3.8 Flash

Google · $0.0551 / 1M tokens

DeepSeek V4

DeepSeek · $0.0613 / 1M tokens

The full list is in the model catalogue and the use-case collections.

Roo Code + OpenLux questions

Does Roo Code work with OpenLux?

Yes. Roo Code exposes an OpenAI Compatible provider, and OpenLux implements the OpenAI Chat Completions API through https://openlux.lol/v1.

Does Roo Code require native tool calling?

Yes. Roo Code uses OpenAI-compatible native tool calling rather than an XML fallback. Choose an OpenLux model tagged Tools and confirm tool calls execute before relying on it for agent work.

Which Roo Code tasks suit cheaper models?

Use Gemini 3.8 Flash or DeepSeek V4 for narrow edits, search and routine explanations when they pass the same task tests. Escalate complex debugging and cross-file changes to GPT-6 Astra or Claude Sonnet 4.5.

How can I track Roo Code spend?

Use a dedicated OpenLux key, review model and cost in the per-request console log, and run a representative task set through the LLM API cost calculator before setting a team budget.

How much will this cost?

Estimate Roo Code API cost

Turn representative IDE tasks into a monthly token and retry budget.

Open the cost guide

Add AI review to the pull-request loop

Build a bounded code reviewer

Use the same OpenAI-compatible endpoint to review pull-request diffs, publish findings and estimate the review budget.

Open the AI code review guide