Platform
Call Logs
Trace LLM outcomes, model selection, latency, cost, retries, and business-safe errors.
What this menu does
Provide read-only operational evidence for AI processing without exposing raw prompts or sensitive model responses.
Where to find it
Platform Admin navigation: LLM Management → Call Logs.
When to use it
- Investigate a failed or slow LLM call.
- Filter calls by firm, provider, model, capability, tier, result, or date.
- Correlate a call with request and trace identifiers.
Before you begin
- Sign in to Platform Admin.
- The account must be allowed to view this menu.
Page regions
- 01Reference and date filters
- 02Call list
- 03Result, token, latency, and error columns
- 04Read-only call detail
- 05Request/trace identifiers
- 06Pagination
Supported operations
Filter call logs
- 1
Choose firm, provider, model, capability, AI tier, and result as needed.
- 2
Set the date range.
- 3
Review time, tokens, latency, cost, retries, and error code.
- 4
Use pagination for older calls.
The list shows matching calls only.
Review call details
- 1
Select Details on a call.
- 2
Review request ID, trace ID, source module/object, provider, model, capability, tier, result, cost, latency, retries, and error.
- 3
Use the identifiers to correlate with support or audit evidence.
- 4
Close the read-only detail.
The operator gains trace context without raw prompt fields.
Business rules and fields
- The page is read-only.
- Raw prompts and model responses are not displayed.
- A call error code is evidence of that call's outcome, not automatically the root cause of a larger incident.
States you may see
What happens next
- Users can trace AI behavior and preserve request identifiers for follow-up without exposing sensitive content.
Common problems
The menu or action is missing.
The account does not have the required platform access or the item is not available in the current state.
Ask a platform administrator to verify the account role and the record status.
A save or load action fails.
The session may have expired, validation may have failed, or the service is temporarily unavailable.
Keep the shown request identifier if present, refresh once, correct highlighted fields, and retry. Escalate with the request identifier if it repeats.