Module transform 

Generic inference request transformation.

Plugins register InferenceRequestTransform implementations to modify the message list and set request-level hints before inference. The core loop applies all registered transforms without knowing their semantics.

Structs

InferenceTransformOutput

Output of an inference request transform.

Traits

InferenceRequestTransform

Trait for plugins that transform the inference request before it is sent to the LLM. The core loop calls all registered transforms in order, piping the output messages of one into the input of the next.

Functions

apply_request_transforms

Apply a chain of transforms, piping messages through each in order.