Skip to main content
Each LLM completion can report token counts via interfaces.LLMUsage on LLMResponse.Usage. OpenAI, Anthropic, and Gemini clients populate the fields when the provider returns them.

​
Usage fields

FieldDescription
PromptTokensInput tokens
CompletionTokensOutput tokens
TotalTokensSum when reported by provider
CachedPromptTokensPrompt cache hits (when supported)
ReasoningTokensExtended thinking tokens (when supported)

​
Run

AgentRunResult.LLMUsage is the sum across all LLM calls in that run — including tool rounds. Use it for cost estimates, quotas, and logging: 
result, err := a.Run(ctx, prompt, nil)
if err != nil {
    return err
}

if result.LLMUsage != nil {
    fmt.Printf("prompt=%d completion=%d total=%d\n",
        result.LLMUsage.PromptTokens,
        result.LLMUsage.CompletionTokens,
        result.LLMUsage.TotalTokens,
    )
}

​
Stream

The same aggregate appears on LLMUsage in the AgentEventTypeRunFinished event — Result is an *AgentRunResult: 
for ev := range eventCh {
    if ev.Type() != agent.AgentEventTypeRunFinished {
        continue
    }
    finished, ok := ev.(*agent.AgentRunFinishedEvent)
    if !ok || finished.Result == nil || finished.Result.LLMUsage == nil {
        continue
    }
    u := finished.Result.LLMUsage
    fmt.Printf("total tokens: %d\n", u.TotalTokens)
}
OpenAI streaming with include_usage surfaces totals on RUN_FINISHED. Example: Stream.

​
RunAsync

Token usage is on the final AgentRunAsyncResult when the run completes — same aggregate as Run.

​
Per-request usage

For per-LLM-call breakdowns, use AfterLLMHook and inspect in.Response.Usage on each iteration.

​
Example

Simple Agent

SHOW_LLM_USAGE footer

Reasoning

ReasoningTokens from extended thinking

Streaming

Usage on RUN_FINISHED events

LLM Providers

Provider-specific usage reporting

Observability

Export usage as metrics