Evaluations

Evaluations

Every evaluator you add is a column in the trace table. A column has three choices that decide what the judge reads and when it runs: a level, an execution mode, and a prompt style (basic rubric or advanced template).

Levels

LevelOne result perThe judge reads
Conversationthreadthe whole transcript, turn by turn
Messageturnthat turn’s user request and agent answer
Stepspan (tool call, generation, …)that step’s input and output

Batch vs sequential

Batch (default) grades every item independently — nothing carries over between items, so two grades of the same item are directly comparable.

Sequential grades items in order as one running conversation with the judge: the rubric is the system prompt, then item → verdict → item → verdict. A step column chains the steps within a message; a message column chains the turns of a thread. Each item is graded knowing what came before it — and the previous result of the same metric is available for continuity.

Sequential grading is incremental: when new turns arrive, only the new turns are graded and appended to the judge’s conversation. Re-running a column from the UI re-grades everything from the start. Conversation-level columns have a single item, so the mode makes no difference there.

Advanced templates and sequential mode

An advanced prompt (any prompt containing @VARIABLES like @HISTORY or @CURRENT_STEP) controls its own context: the resolved template is the whole prompt, and nothing is added around it.

Because of that, sequential works differently for advanced columns: there is no running conversation with the judge. Chaining happens only through @METRIC_PREVIOUS_RESULT, which resolves to the previous item’s result of the same metric — put it in your template where you want the judge to see it. If your template doesn’t reference it, an advanced sequential column behaves exactly like batch.

Output types

score (0–1 with a pass/fail threshold), number, boolean, text, or json — your own schema, built in the Add Column form. Enum fields are enforced on the model, and a numeric score field in your schema drives the pass/fail verdict via the threshold.

Further reading

New to this? LLM evaluation: metrics, methods and tools covers the concepts behind this page — offline versus online evaluation, when a deterministic check beats a judge, why the evaluation level decides what you can catch, and how to calibrate a judge before you let it gate a release.