On This Page

Home / LLM Observability/LLM Telemetry Use Cases in Cribl

LLM Telemetry Use Cases in Cribl

With large language model (LLM) application telemetry flowing into Cribl Search and Cribl Stream, you can explore and visualize LLM telemetry, and control where the data goes and how it is shaped.

The following topics are vendor-agnostic patterns you can map to the field names and span types your instrumentation exposes. For background on common semantic conventions, see resources such as OpenInference.

Explore the following use cases:

Typical Field Names

Exact field names vary by instrumentation. The examples in the guides use generic names like total_tokens, model, and estimated_cost_usd for clarity–substitute the fields available in your telemetry.

ConceptExample field names
Model namellm.request.model, model, request.model
Prompt tokensllm.usage.prompt_tokens, prompt_tokens
Completion tokensllm.usage.completion_tokens, completion_tokens
Total tokensllm.usage.total_tokens, total_tokens
Per-request costllm.request.cost, llm.usage.cost, cost_usd
Span/operation typespan.kind, component, operation, llm.span_type
Environment/deploymentdeployment.environment, env, environment
User identifiersuser.id, tenant.id, customer_id, account_id