AI Gateway has a unified fast mode abstraction, now in beta.
You can now request fast mode the same way for every model on AI Gateway. Set speed to fast
, and the gateway serves the fast tier when it's available and falls back to standard speed when it isn't.
Fast mode trades a higher per-token cost for lower latency or higher throughput. You get it wherever it is available, without pinning a provider or wiring anything up yourself.
Set speed
to fast
under providerOptions.gateway
to upgrade the model to its fast serving path when one is routable:
You can also address a fast variant directly with its fast slug, such as anthropic/claude-opus-5-fast
, which is the same as setting speed: 'fast'
on the base model.
Use the speed
option when you want to stay on the base model ID and fall back to standard speed. Use a fast slug when you want to name the fast variant directly, for example in a gateway.models
fallback list. The speed
option works across every AI Gateway API format.
Fast mode is available for a growing set of models. Find them in the models list with a lightning bolt next to the slug. Requesting fast mode on a model with no fast tier has no effect, and the request runs at standard speed. Fast variants of models are usually more expensive than the base model: check pricing for more details.
In Claude Code, toggle fast mode with /fast
for Anthropic Opus models after setup. For other agents, select a fast slug in your model configuration, such as anthropic/claude-opus-5-fast
.
For more information about fast mode, read the documentation.