Skip to content

Easy8 AI Basic with self-hosted AI

Use this guide for Easy8 AI Basic with an LLM that runs on your own infrastructure. Basic needs no ChromaDB and no embeddings service.

Before you start, read Prerequisites.

Supported LLM

Easy8 AI Basic supports gemma-4-26b served through vLLM. The endpoint in LLM_URL must expose this model.

If your vLLM endpoint exposes a deployment-specific model id, map it with VLLM_MODEL_ALIASES as described in Custom vLLM served model aliases.

Environment configuration

Set the variables from Common configuration, then add the LLM endpoint:

LLM_URL=https://llm.example.com

If your vLLM endpoint or the reverse proxy in front of it requires an API key, also set VLLM_API_KEY; Chain sends it as Authorization: Bearer <key>. The vLLM deployment example uses such a proxy.

LLM_URL can contain several comma-separated endpoints. If the endpoint uses HTTPS with a certificate from your own CA, configure the CA bundle as described in Custom CA certificates.

Installation steps

  1. If you are preparing a new GPU host, follow Self-hosted AI server setup.
  2. Deploy the LLM with the vLLM deployment example. The endpoint must be reachable from the chain container.
  3. Add LLM_URL and the common variables to the Easy8 .env file.
  4. Check that the package includes docker-compose_easy_ai.yml.

  5. Install and start Easy8:

    docker compose run --rm app install
docker compose up -d

  6. Check that app, chain, celery and redis are running.

  7. Add the reverse proxy rule from Nginx configuration.

  8. Open both public Chain URLs:

    • https://your.easy.host/ai/chain/sys/monitoring.json
    • https://your.easy.host/ai/chain/api/info

    The info endpoint must report the vLLM provider and model. See Verification and troubleshooting.