What is a Session?
A session groups the turns of a single conversation between a user and your AI product. It contains one or more inference results (input/output pairs) and can be evaluated as a whole using conversational metrics. Sessions must be created explicitly withsessions.create() before logging inference results and evaluations. This gives you full control over session lifecycle and grouping.
You can create, view, and manage sessions programmatically using the Galtea SDK.
SDK Integration
Session Service SDK
Manage sessions using the Python SDK
Session Properties
The custom ID associated with the session.
This is usually a client-side generated ID that identifies the conversation session between a user and a LLM app.
The version associated with the session.
The test case associated with the session. This is typically used for non-production evaluations.
A creation-time parameter that indicates whether the session is a production session. When set to
True, no test_case_id is required and the provided context is used directly. When False (the default), a test_case_id must be provided and the context is derived from the test case.
This is not a stored field on the session. Whether a session is a production session can be inferred at read time by checking whether
test_case_id is null.The context provided to the AI product during the session or ground-truth related context. For production sessions, this value is taken from the request. For non-production sessions, it is derived from the linked test case.
Additional custom metadata about the session. This can be used to store any relevant information that doesn’t fit into the other fields.
The current status of the session. Possible values:
PENDING: The session has been created but has not yet received all inference results.COMPLETED: The session has finished successfully.FAILED: The session encountered an error during processing.
Any error message associated with the session. This is useful for tracking failed sessions or sessions that encountered issues.
Related
Evaluation
Learn how evaluations score inference results within a session.
Inference Result
Understand the input/output pairs that make up a session.
Simulating Conversations
Tutorial on running multi-turn conversation simulations.