A reference from Weave

Model routing and gateways glossary

Learn how applications choose models, manage providers, and recover from failures. Understand the tradeoffs between answer quality, response time, and cost.

All terms

120 terms

  • Active-active routing

    Active-active routing is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Active-passive routing

    Active-passive routing is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • AI gateway

    AI gateway is a model-routing or gateway concept used to manage interface consistency for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • API gateway for LLMs

    API gateway for LLMs is a model-routing or gateway concept used to manage interface consistency for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Authentication proxy

    Authentication proxy is a model-routing or gateway concept used to manage interface consistency for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Backpressure

    Backpressure is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Blue-green routing

    Blue-green routing is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Budget policy

    Budget policy is a model-routing or gateway concept used to manage policy enforcement for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Canary routing

    Canary routing is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Capability routing

    Capability routing is a model-routing or gateway concept used to manage operational visibility for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Classifier-based routing

    Classifier-based routing is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Compliance routing

    Compliance routing is a model-routing or gateway concept used to manage policy enforcement for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Concurrency limit

    Concurrency limit is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Configuration drift

    Configuration drift is a model-routing or gateway concept used to manage operational visibility for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Configuration reload

    Configuration reload is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Configuration validation

    Configuration validation is a model-routing or gateway concept used to manage operational visibility for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Connect timeout

    Connect timeout is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Consistent hashing routing

    Consistent hashing routing is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Context length routing

    Context length routing is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Credential injection

    Credential injection is a model-routing or gateway concept used to manage interface consistency for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Cross-region failover

    Cross-region failover is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Data residency routing

    Data residency routing is a model-routing or gateway concept used to manage policy enforcement for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Deadline propagation

    Deadline propagation is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Dry-run routing

    Dry-run routing is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Duplicate request protection

    Duplicate request protection is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Dynamic routing

    Dynamic routing is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Endpoint discovery

    Endpoint discovery is a model-routing or gateway concept used to manage operational visibility for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Endpoint health

    Endpoint health is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Endpoint registry

    Endpoint registry is a model-routing or gateway concept used to manage operational visibility for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Error normalization

    Error normalization is a model-routing or gateway concept used to manage interface consistency for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Exponential backoff

    Exponential backoff is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Failover routing

    Failover routing is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Fairness policy

    Fairness policy is a model-routing or gateway concept used to manage policy enforcement for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Fallback chain

    Fallback chain is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Fallback reason

    Fallback reason is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Fallback threshold

    Fallback threshold is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Gateway adapter

    Gateway adapter is a model-routing or gateway concept used to manage interface consistency for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Gateway configuration

    Gateway configuration is a model-routing or gateway concept used to manage interface consistency for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Gateway endpoint

    Gateway endpoint is a model-routing or gateway concept used to manage interface consistency for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Gateway metrics

    Gateway metrics is a model-routing or gateway concept used to manage interface consistency for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Gateway request lifecycle

    The gateway request lifecycle is the ordered set of stages an AI request passes through before a response reaches the application. It commonly includes authentication, admission, policy evaluation, routing, provider execution, retries or fallback, response handling, and usage recording.

    Model routing and gateways

  • Graceful degradation

    Graceful degradation is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Idempotency key

    Idempotency key is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Least-loaded routing

    Least-loaded routing is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Load shedding

    Load shedding is a model-routing or gateway concept used to manage reliable request governance for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Metadata normalization

    Metadata normalization is a model-routing or gateway concept used to manage interface consistency for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Modality routing

    Modality routing is a model-routing or gateway concept used to manage traffic selection for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Model allowlist

    Model allowlist is a model-routing or gateway concept used to manage policy enforcement for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Model capability matrix

    Model capability matrix is a model-routing or gateway concept used to manage operational visibility for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • Model denylist

    Model denylist is a model-routing or gateway concept used to manage policy enforcement for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways