All Classes and Interfaces
Class
Description
Utility to serialize LLM attachment data in a braintrust-friendly manner.
Provides access to the distributed trace spans for a single eval task stored in Braintrust.
Main entry point for the Braintrust SDK.
Deprecated.
Request body for BTQL queries.
Response from a BTQL query.
Represents a Braintrust function (scorer, prompt, tool, or task).
Request body for invoking a function.
Configuration for Braintrust SDK with sane defaults.
Used to identify the braintrust parent for spans and experiments.
Centralized ObjectMapper for the Braintrust SDK.
Provides the necessary API calls for the Braintrust SDK.
Response from a
POST /btql query.Real-time indexing state for a BTQL query.
Load LLM prompts from the Braintrust API
Custom span processor that enriches spans with Braintrust-specific attributes.
Parent context for spans (project or experiment).
Main entry point for Braintrust tracing setup.
Represents a parsed parent with type and ID.
Thrown when the BTQL endpoint returns HTTP 429 (Too Many Requests).
A single structured classification produced by a
Classifier.A classifier categorizes and labels eval outputs, producing zero or more structured
Classification items.Datasets define the cases for evals.
A dataset loaded externally from Braintrust using paginated API fetches
A single row in a dataset.
Remote Eval Dev Server
An evaluation framework for testing AI models.
Builder for creating evaluations with fluent API.
Deprecated.
Request body for POST /eval endpoint
Dataset specification - supports inline data, by name, or by ID
Individual evaluation case data
Function ID specification (multiple formats supported)
Remote scorer specification
Response body for POST /eval endpoint
Summary statistics for a scorer
Results of all eval cases of an experiment.
Exception thrown when login to Braintrust fails.
Generic pointer to an object in braintrust
Definition of a named parameter that can be configured from the Braintrust Playground UI.
Holds the merged parameter values for a single eval run, along with the parameter definitions.
Represents a remote evaluator that can be exposed via the dev server.
Individual metric value assigned by a scorer.
A scorer evaluates the result of a task against a dataset case, producing a score between 0
(inclusive) and 1 (inclusive).
A scorer that invokes a remote Braintrust function to compute scores.
Represents span components for distributed tracing in object format.
Row IDs for linking spans within a trace.
A task function that runs against a single dataset case and produces an output.
Result from a single task run.
A classifier that receives access to the full distributed trace of the task that was evaluated.
A scorer that receives access to the full distributed trace of the task that was evaluated.