Test an HTTP interactions endpoint on a free web service. Run a Gateway bot on a paid background worker.
Each instance has a fixed monthly price, prorated by the second.
Capability | Render | |||
|---|---|---|---|---|
How Discord bots run | Paid worker for Gateway bots or web service for HTTP | Long-running containers | Worker dynos, restarted daily by default | Fly Machine built from a Dockerfile |
Free tier | Free HTTP testing. Gateway workers need paid compute | $5 trial for 30 days, then $1/month credit | None. Eco: $5/month for 1,000 hours | No free tier. Personal bills under $5 waived |
Always-on, 512 MB | $7/month (0.5 CPU) | Up to $15/month at full use (0.5 vCPU) | $7/month (Basic dyno) | $3.69/month (shared-cpu-1x) |
Stack as code | One render.yaml, auto-synced on push | Project file, applied with CLI or CI | app.json per app, read only at creation | fly.toml per app. No database definitions |
Managed Postgres and Redis | Postgres from $6 with PITR, plus Key Value | Maintained databases. HA and Postgres PITR opt-in | Postgres and Key-Value. No rollback on Essential | Managed Postgres from $38, HA included. Upstash Redis |
Scheduled tasks | Cron jobs | Cron jobs | Scheduler add-on | Supercronic or Cron Manager |
Persistent disks | SSDs that grow with no downtime (single instance) | Volumes (no replicas) | None. Files reset when dynos restart | Volumes, one per Machine, not replicated |
Competitor details as of November 2026. Instance prices are compute only. Fly.io prices are for US East.
Choose a worker or web service based on how your bot receives events. Define the bot and its data in a render.yaml Blueprint.
Maintain the outbound Discord connection in a continuous paid worker. No public HTTP endpoint is required.
Expose your interaction handler over HTTPS. Validate Discord request signatures and acknowledge interactions within its deadline.
Store bot tokens and API credentials as secret environment variables. Rotate compromised tokens through the Discord developer portal.
Use Postgres for guild settings, moderation history, or application data that must survive restarts.
Move long processing to a worker and shared queue, so your interaction endpoint can respond promptly.
Use a Discord library that handles heartbeats and reconnects. Coordinate shards and event processing before adding instances.
Create a Background Worker and connect the repository containing your Node.js or Python bot.
Set the dependency-install command, production start command, and bot token as a secret environment variable.
Deploy the worker, check logs, and confirm the bot connects to Discord with the required intents. Test reconnect behavior before inviting it to production servers.