Integration guide

Point Continue at OpenLux

Continue is configured with a single YAML file that lists chat, autocomplete and embedding models. Every entry can point at OpenLux, so the IDE assistant runs on discounted rates with one key.

Create an API key Read the docs

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

Set it up in four steps

Create an OpenLux key

Create the key in the console. Continue uses it for chat, edits and autocomplete, so one key covers the whole assistant.

Edit config.yaml

Open ~/.continue/config.yaml and add models with provider: openai and the OpenLux base URL in apiBase.

Assign roles to models

Give the chat role a strong model, leave autocomplete on a cheap one and add an embeddings entry so codebase indexing also runs through OpenLux.

Index and work normally

Continue indexes the repository through the embedding model you configured, then answers with whatever chat model is selected.

Configuration

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

~/.continue/config.yaml

models:
  - name: OpenLux GPT-6 Astra
    provider: openai
    model: gpt-6-astra
    apiBase: https://openlux.lol/v1
    apiKey: ${OPENLUX_API_KEY}
    roles: [chat, edit]
  - name: OpenLux DeepSeek V4
    provider: openai
    model: deepseek-v4
    apiBase: https://openlux.lol/v1
    apiKey: ${OPENLUX_API_KEY}
    roles: [autocomplete]
  - name: OpenLux BGE-M3
    provider: openai
    model: bge-m3
    apiBase: https://openlux.lol/v1
    apiKey: ${OPENLUX_API_KEY}
    roles: [embed]

Models that fit Continue

GPT-6 Astra

OpenAI · $0.3677 / 1M tokens

Claude Sonnet 4.5

Anthropic · $0.2647 / 1M tokens

DeepSeek V4

DeepSeek · $0.0613 / 1M tokens

BGE-M3

AIGC · $0.0035 / 1M tokens

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

Continue + OpenLux questions

Does Continue work with OpenAI-compatible endpoints?

Yes. Each model entry accepts provider, apiBase and apiKey, so any compatible endpoint can be used for chat, autocomplete or embeddings.

Can different roles use different models?

Yes, and it is worth doing: an expensive flagship for chat and edits, a cheap fast model for autocomplete — the inline cost keeps both honest.

Do embeddings need a separate vendor?

No. BGE-M3 is in the OpenLux catalogue and is metered at $0.0035 per million tokens, so codebase indexing runs on the same key.

How do keys stay out of the config file?

Reference an environment variable in config.yaml and export it in your shell profile instead of pasting the secret into the file itself.

Connect Continue to OpenLux today

One key covers every model in this guide and the rest of the catalogue, metered in USD with free trial credit to start.

Create an API key See pricing