Digitorn
Digitorn
All terms
Cost & ops

Brain fallback

A second LLM provider the runtime swaps to when the primary returns an error or runs out of credit.

also known as: fallback provider
In depth

Brain fallback is what saves your Saturday afternoon when an Anthropic balance hits zero. The YAML declares a primary brain and a fallback brain, the runtime watches for 402, 429, and 'insufficient credit' errors and silently swaps to the fallback for the next turn. The agent keeps running, the user does not notice. Next turn retries the primary first, so you do not stay on the fallback longer than needed.

What it looks like in practice

YAML
1brain:2  provider: anthropic3  model: claude-sonnet-4-64  fallback:5    provider: deepseek6    model: deepseek-chat
Related concepts
Newsletter

Get the next post in your inbox.

Engineering notes from the Digitorn team. No marketing, no launch announcements, no "10 prompts that will change your life". Just the things we write that we'd want to read.

One-click unsubscribe. We never share your address. Powered by our own infrastructure, not a tracker.

More in Cost & ops

Audit log/glossary/audit-logCost cap/glossary/cost-capModel routing/glossary/model-routingPII redaction/glossary/pii-redactionRate limiting/glossary/rate-limitingTurn cap/glossary/turn-cap