Cost & ops
Canary Release
Also known as canary deployment, canary rollout
A canary release exposes a change to a small, deliberately chosen slice of traffic first.
A bad model version, prompt, or index build then shows up against a bounded audience, not everyone.
The cohort is chosen for how much damage a bad version could do, not for convenience.
The ramp to full traffic follows a written schedule with abort criteria set before anyone sees the change.
Watching only for hard errors misses the failure that matters most: a quality regression with no exception.
← Back to the full glossaryRelated terms
Output Drift Output drift is a gradual change in a live system's output, with no release behind it. Model Deprecation Model deprecation is a provider retiring a model version on its own schedule. Shadow Deployment A shadow deployment runs a candidate release against real production traffic, offline, before release.
More in Cost & ops
Inference Cost Inference cost is what it costs to run a trained model on a request. Notebook-to-Production Notebook-to-production is the gap between a working prototype and a system that runs reliably. Acceptance Rate Acceptance rate is the share of an agent's proposed changes a human reviewer approves.