Documentation / Connect

Connect → LibreChat

Use AxForge with LibreChat

LibreChat is an open-source chat app you host yourself. AxForge goes in as a custom endpoint in its librechat.yaml, and our models show in its menu under AxForge.

Before you start
  • An AxForge API key. Get one (a minute).
  • LibreChat running with Docker, as in its own guide. This guide was tried with LibreChat v0.8.7 on 2026-10-17.

Set it up

  1. Add AxForge in librechat.yaml

    In LibreChat's folder, create librechat.yaml:

    version: 1.3.13
    cache: true
    
    endpoints:
      custom:
        - name: 'AxForge'
          apiKey: '${AXFORGE_API_KEY}'
          baseURL: 'https://api.axforge.ai/v1'
          models:
            default:
              - 'qwen3.8-27b'
              - 'glm-5.3-flash'
              - 'mimo-v2.6-flash'
            fetch: false
          titleConvo: true
          titleModel: 'mimo-v2.6-flash'
          modelDisplayLabel: 'AxForge'

    fetch: false keeps the menu to these three chat models. titleModel names each new chat with the fastest one.

  2. Put your key in .env

    Add one line to LibreChat's .env; ${AXFORGE_API_KEY} above reads it from there:

    AXFORGE_API_KEY=orx_live_your_key_here
  3. Let Docker see the file, and restart

    LibreChat's Docker setup reads librechat.yaml through an override file next to its docker-compose.yml:

    services:
      api:
        volumes:
          - type: bind
            source: ./librechat.yaml
            target: /app/librechat.yaml
    docker compose down
    docker compose up -d
  4. Pick the model

    Open LibreChat (sign up: the first account is a local one), open the model menu at the top, choose AxForge and then qwen3.8-27b.

    LibreChat's model menu with AxForge open and qwen3.8-27b marked, beside glm-5.3-flash and mimo-v2.6-flash
  5. Ask something

    A LibreChat conversation with a question and the answer from AxForge

The settings

Key in librechat.yamlValue
baseURLhttps://api.axforge.ai/v1
apiKey${AXFORGE_API_KEY}, from .env
models.defaultqwen3.8-27b, glm-5.3-flash, mimo-v2.6-flash
titleModelmimo-v2.6-flash

Which model for what: qwen3.8-27b is the everyday default and reads images, glm-5.3-flash thinks before it answers, mimo-v2.6-flash is the fastest.

Check that it works

AxForge is in the model menu with our three models, and a question gets an answer. The calls show in the console under API Keys → Recent calls within a minute.

When it does not work

Not sure which it is? The connection doctor makes seven small checks on your key and names the fix: irm https://dev.axforge.ai/assets/axforge-doctor.ps1 | iex in PowerShell, curl -fsSL https://dev.axforge.ai/assets/axforge-doctor.sh | bash on macOS and Linux.

What you seeWhat it meansFix
No AxForge in the menuLibreChat did not read librechat.yamlcheck the override file, then docker compose down and up -d
"Invalid API key. Create one at https://console.axforge.ai/keys."AXFORGE_API_KEY is missing from .env, or the key was refusedadd the line, the key whole, and restart

Test a call says in one click whether a problem is your key, your balance or us.

Ask on the forum Markdown For AI Updated 2026-11-06

Anything unclear on this page?

Ask on the forum. The answer helps the next person too.

Ask about this page