Skip to content

Render Deployment

Deploy LLM Council to Render with the free tier for evaluation, or paid tier for production.

Important: Free Tier Limitations

Cold Start Warning

Render Free tier spins down after 15 minutes of inactivity. This causes:

  • 30-60 second cold starts
  • Potential webhook timeouts (n8n, Make, Zapier)
  • Unreliable for production use

For webhook integrations, use Railway or Render paid tier.

One-Click Deploy

Deploy to Render

Manual Setup

1. Fork the Repository

Fork amiable-dev/llm-council to your GitHub account.

2. Create New Web Service

  1. Go to Render Dashboard
  2. Click "New" > "Web Service"
  3. Connect your forked repository

3. Configure Build Settings

Setting Value
Name llm-council
Environment Docker
Dockerfile Path deploy/railway/Dockerfile
Instance Type Free (evaluation) or Starter ($7/mo)

4. Add Environment Variables

In the Environment section, add:

Key Value
OPENROUTER_API_KEY Your OpenRouter API key
LLM_COUNCIL_API_TOKEN Your secure API token

Generate a Secure Token

openssl rand -hex 16

5. Configure Health Check

Render auto-detects from render.yaml, but verify:

Setting Value
Health Check Path /health

6. Deploy

Click "Create Web Service". Render will build and deploy your instance.

Blueprint Deployment

Alternatively, use the included render.yaml blueprint:

  1. Fork the repository
  2. Go to Render Dashboard > "New" > "Blueprint"
  3. Select your forked repo
  4. Render will read render.yaml and configure automatically
# render.yaml (already in repo)
services:
  - type: web
    name: llm-council
    runtime: docker
    dockerfilePath: deploy/railway/Dockerfile
    healthCheckPath: /health
    autoDeploy: false

Verify Deployment

Health Check

curl https://your-app.onrender.com/health
# {"status":"ok","service":"llm-council-local"}

First Request Cold Start

If the service is cold, the first request may take 30-60 seconds.

API Test

curl -X POST https://your-app.onrender.com/v1/council/run \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "What is the capital of France?"}'

Keeping Service Warm

For Free tier with webhook needs, consider these workarounds:

Option 1: External Ping Service

Use a free uptime monitor to ping /health every 10 minutes:

Option 2: Scheduled Cron Job

Add a GitHub Action to ping your service:

# .github/workflows/keep-warm.yml
name: Keep Render Warm
on:
  schedule:
    - cron: '*/10 * * * *'  # Every 10 minutes
jobs:
  ping:
    runs-on: ubuntu-latest
    steps:
      - run: curl -f ${{ secrets.RENDER_URL }}/health

Option 3: Upgrade to Paid Tier

For $7/mo, Render Starter tier eliminates cold starts.

Troubleshooting

Build Fails

Check Dockerfile path is correct: deploy/railway/Dockerfile

Service Never Starts

  1. Check deployment logs for errors
  2. Verify OPENROUTER_API_KEY is set
  3. Ensure Dockerfile builds locally: 
    docker build -f deploy/railway/Dockerfile -t llm-council-test .

Webhooks Timeout

This is the cold start problem. Solutions:

  1. Switch to Railway (recommended)
  2. Upgrade to Render paid tier
  3. Use keep-warm ping service

Cost Comparison

Tier Monthly Cost Cold Starts Webhook Ready
Free $0 Yes (15min) No
Starter $7 No Yes
Standard $25 No Yes

Next Steps