For the complete documentation index, see llms.txt.
Skip to main content
Version: 8.10

Type Alias: AgentInstanceHistoryItemMetricsRequest

type AgentInstanceHistoryItemMetricsRequest = object;

Per-call token and latency metrics for an ASSISTANT history item, as submitted on a create/update request. All fields are optional: omit a field the caller has no value for rather than sending it as an explicit null.

Properties​

cacheCreationTokenCount?​

optional cacheCreationTokenCount?: number | null;

Cache-creation tokens consumed by this LLM call. Null when not provided.

cacheReadTokenCount?​

optional cacheReadTokenCount?: number | null;

Cache-read tokens consumed by this LLM call. Null when not provided.

durationMs?​

optional durationMs?: number | null;

Wall-clock duration of the LLM call in milliseconds. Null when not provided.

inputTokens?​

optional inputTokens?: number | null;

Input tokens consumed by this LLM call. Null when not provided.

outputTokens?​

optional outputTokens?: number | null;

Output tokens produced by this LLM call. Null when not provided.

reasoningTokenCount?​

optional reasoningTokenCount?: number | null;

Reasoning tokens consumed by this LLM call. Null when not provided.