InferenceAttempt

Type object

Shape

Field	Type	Description
idrequired	InferenceAttemptId	InferenceAttemptId - a client-minted UUID v4, never derived from content or location.
operation_idrequired	string
chat_id	ChatId or null
assistant_message_id	MessageId or null
surfacerequired	AiSurface
intentrequired	AiIntent
providerrequired	AiProviderId	Extensible provider token such as openai, ollama, or koru.
requested_modelrequired	string
resolved_model	string or null
requested_reasoningrequired	ReasoningChoice	A provider's reasoning setting - `ai.db`'s nine-token CHECK.
effective_reasoning	ReasoningChoice or null
prompt_template_versionrequired	string
statusrequired	AttemptStatus
queued_atrequired	Timestamp	Epoch milliseconds, UTC. Signed, so instants before 1970 are representable.
context_started_at	Timestamp or null
context_completed_at	Timestamp or null
provider_started_at	Timestamp or null
first_token_at	Timestamp or null
completed_at	Timestamp or null
request_payload_bytes	integer (uint64, >= 0) or null
web_search_requested	boolean	Default `false`
web_search_calls	integer (uint32, >= 0)	Default `0`
context_payload_bytesrequired	integer (uint64, >= 0)
message_countrequired	integer (uint32, >= 0)
context_document_countrequired	integer (uint32, >= 0)
context_message_block_countrequired	integer (uint32, >= 0)
usagerequired	AiUsage
context_used_tokens	integer (uint64, >= 0) or null	Latest active-context pair reported by this attempt's provider. It is distinct from response usage and absent for older attempts. Default `null`
context_limit_tokens	integer (uint64, >= 0) or null	Default `null`
stop_reason	string or null
error_class	AiErrorClass or null
provider_response_id	string or null

Example

{
  "id": "00000000-0000-4000-8000-000000000000",
  "operation_id": "string",
  "chat_id": "00000000-0000-4000-8000-000000000000",
  "assistant_message_id": "00000000-0000-4000-8000-000000000000",
  "surface": "standalone",
  "intent": "chat",
  "provider": "string",
  "requested_model": "string",
  "resolved_model": "string",
  "requested_reasoning": "default",
  "effective_reasoning": "default",
  "prompt_template_version": "string",
  "status": "pending",
  "queued_at": 0,
  "context_started_at": 0,
  "context_completed_at": 0,
  "provider_started_at": 0,
  "first_token_at": 0,
  "completed_at": 0,
  "request_payload_bytes": 0,
  "web_search_requested": false,
  "web_search_calls": 0,
  "context_payload_bytes": 0,
  "message_count": 0,
  "context_document_count": 0,
  "context_message_block_count": 0,
  "usage": {
    "prompt_tokens": 0,
    "cached_prompt_tokens": 0,
    "completion_tokens": 0,
    "reasoning_tokens": 0,
    "total_tokens": 0,
    "source": "provider_reported"
  },
  "context_used_tokens": 0,
  "context_limit_tokens": 0,
  "stop_reason": "string",
  "error_class": "configuration",
  "provider_response_id": "string"
}

The contract has no example values, so example bodies are generated from each schema: every field is shown with a placeholder of its type.

Generated from contract version 1. Last updated November 12, 2026.