Responses
Compact a Response
Compacts a conversation by summarizing older items to free up context window space. Returns the next context window: a compaction item containing the generated summary, followed by the most recent items verbatim.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.responses.compact(input=[
{
"content": "Tell me about machine learning",
"role": "user",
"type": "message",
},
{
"content": "Machine learning is a subset of AI...",
"role": "assistant",
"type": "message",
},
], model="openai/gpt-6-astra")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.responses.compact({
input: [
{
content: "Tell me about machine learning",
role: "user",
type: "message",
},
{
content: "Machine learning is a subset of AI...",
role: "assistant",
type: "message",
},
],
model: "openai/gpt-6-astra",
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"input": Union[str, List[Input2]], # optional
"instructions": Optional[str],
"model": Optional[str],
"previous_response_id": Optional[str],
"prompt_cache_key": Optional[str],
}
{
input?: string | Input2[];
instructions?: string;
model?: string;
previousResponseId?: string;
promptCacheKey?: string;
}
Show Response
Show Response
{
"created_at": int,
"id": str,
"object": Literal["response.compaction"],
"output": List[Any],
"usage": {
"input_cost": Optional[float],
"input_tokens": int,
"input_tokens_details": {
"cache_creation_1h_tokens": Optional[int],
"cache_creation_5m_tokens": Optional[int],
"cache_creation_tokens": int,
"cache_write_tokens": int,
"cache_write_tokens_details": { # optional
"ttl_12h_tokens": int,
"ttl_30m_tokens": int,
},
"cached_tokens": int,
},
"output_cost": Optional[float],
"output_tokens": int,
"output_tokens_details": {
"reasoning_tokens": int,
},
"server_tool_use": { # optional
"advisor_requests": Optional[int],
"code_interpreter_sessions": Optional[int],
"datetime_requests": Optional[int],
"fusion_requests": Optional[int],
"image_generation_calls": Optional[int],
"search_models_requests": Optional[int],
"shell_commands": Optional[int],
"subagent_requests": Optional[int],
"tool_search_requests": Optional[int],
"web_fetch_requests": Optional[int],
"web_search_requests": Optional[int],
},
"speed": Optional[Literal["standard", "fast"]],
"total_cost": Optional[float],
"total_tokens": int,
"web_search_requests": Optional[int],
},
}
{
createdAt: number;
id: string;
object: "response.compaction";
output: any[];
usage: {
inputCost?: number;
inputTokens: number;
inputTokensDetails: {
cacheCreation1hTokens?: number;
cacheCreation5mTokens?: number;
cacheCreationTokens: number;
cacheWriteTokens: number;
cacheWriteTokensDetails?: {
ttl12hTokens: number;
ttl30mTokens: number;
};
cachedTokens: number;
};
outputCost?: number;
outputTokens: number;
outputTokensDetails: {
reasoningTokens: number;
};
serverToolUse?: {
advisorRequests?: number;
codeInterpreterSessions?: number;
datetimeRequests?: number;
fusionRequests?: number;
imageGenerationCalls?: number;
searchModelsRequests?: number;
shellCommands?: number;
subagentRequests?: number;
toolSearchRequests?: number;
webFetchRequests?: number;
webSearchRequests?: number;
};
speed?: "standard" | "fast";
totalCost?: number;
totalTokens: number;
webSearchRequests?: number;
};
}
Show Variant types
Show Variant types
Input2
{
"content": Optional[str],
"id": Optional[str],
"role": Optional[Literal["user", "assistant", "system"]],
"type": Optional[Literal["message", "function_call", "function_call_output", "item_reference", "reasoning", "custom_tool_call", "custom_tool_call_output", "compaction"]],
}
{
content?: string;
id?: string;
role?: "user" | "assistant" | "system";
type?: "message" | "function_call" | "function_call_output" | "item_reference" | "reasoning" | "custom_tool_call" | "custom_tool_call_output" | "compaction";
}
Create a Response
Creates a model response for the given input. Returns a response object or a stream of server-sent events.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.responses.create(input="Summarise the latest deployment")
with res as event_stream:
for event in event_stream:
# handle event
print(event, flush=True)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.responses.create({
input: "Summarise the latest deployment",
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"background": Optional[bool],
"cache": { # optional
"ttl": Optional[int],
"type": str, # required
},
"cache_control": { # optional
"ttl": Optional[Literal["5m", "1h"]],
"type": Literal["ephemeral"], # required
},
"context_management": [{ # optional
"compact_threshold": Optional[int],
"provider_native": Optional[bool],
"type": str, # required
}],
"conversation": { # optional
"id": str, # required
},
"fallback_on_codes": List[int], # optional
"fallbacks": [{ # optional
"model": str, # required
}],
"frequency_penalty": Optional[float],
"guardrails": [{ # optional
"execute_on": Literal["input", "output", "both"], # required
"id": str, # required
"is_guardrail": Optional[bool],
"sample_rate": Optional[float],
"timeout": Optional[int],
}],
"identity": { # optional
"display_name": Optional[str],
"email": Optional[str],
"id": str, # required
"metadata": List[Dict[str, Any]], # optional
"tags": List[str], # optional
},
"include": List[str], # optional
"input": Union[str, List[CreateRouterResponseInput2]], # optional
"instructions": Optional[str],
"integration_id": Optional[str],
"limits": { # optional
"max_cost": Optional[float],
"max_depth": Optional[int],
"max_execution_time": Optional[int],
"max_iterations": Optional[int],
"tool_timeout": Optional[int],
},
"load_balancer": { # optional
"models": [{ # required
"model": str, # required
"weight": float, # required
}],
"type": Literal["weight_based", "round_robin", "latency_based"], # required
},
"max_output_tokens": Optional[int],
"max_tool_calls": Optional[int],
"memory": { # optional
"entity_id": str, # required
},
"metadata": Dict[str, str], # optional
"model": Optional[str],
"parallel_tool_calls": Optional[bool],
"plugins": [{ # optional
"entities": List[str], # optional
"entity_thresholds": Dict[str, float], # optional
"id": Literal["pii_redaction", "response_healing", "trace_scrubbing"], # required
"language": Optional[str],
"mask": List[Literal["all", "system", "input", "output", "metadata", "variables"]], # optional
"on_failure": Optional[Literal["block", "passthrough"]],
"regions": List[str], # optional
"threshold": Optional[float],
}],
"presence_penalty": Optional[float],
"previous_response_id": Optional[str],
"prompt_cache_key": Optional[str],
"prompt_cache_options": { # optional
"mode": Optional[Literal["implicit", "explicit"]],
"ttl": Optional[Literal["30m", "12h"]],
},
"reasoning": { # optional
"context": Optional[Literal["auto", "current_turn", "all_turns"]],
"effort": Optional[Literal["none", "instant", "minimal", "low", "medium", "high", "xhigh", "max"]],
"summary": Optional[Literal["concise", "detailed", "auto"]],
},
"retry": { # optional
"count": int, # required
"on_codes": List[int], # optional
},
"safety_identifier": Optional[str],
"security": { # optional
"mask": List[str], # optional
},
"service_tier": Optional[Literal["auto", "default", "flex", "fast", "ultrafast", "scale", "priority"]],
"speed": Optional[Literal["standard", "fast"]],
"stop_sequences": List[str], # optional
"store": Optional[bool],
"stream": Optional[bool],
"stream_options": { # optional
"include_obfuscation": bool, # required
},
"temperature": Optional[float],
"template_engine": Optional[Literal["text", "jinja", "mustache"]],
"text": { # optional
"format_": { # optional
# one of: PlainText, FormatJSONSchema
"type": Union[Literal["text"], Literal["json_schema"]], # required
"description": Optional[str],
"name": str, # required (variant-specific)
"schema_": Dict[str, Any], # required (variant-specific)
"strict": Optional[bool],
},
"verbosity": Optional[Literal["low", "medium", "high"]],
},
"thinking": { # optional
"budget_tokens": Nullable[int], # required
"display": Optional[Literal["summarized", "omitted", "updates"]],
"thinking_level": Optional[Literal["minimal", "low", "medium", "high"]],
"type": Nullable[ThinkingParamType], # required
},
"thread": { # optional
"id": str, # required
"tags": List[str], # optional
},
"timeout": { # optional
"call_timeout": int, # required
},
"tool_choice": { # optional
# one of: Shorthand, SpecificFunction
"name": str, # required (variant-specific)
"type": Literal["function"], # required (variant-specific)
},
"tools": List[Union[ToolsFunction, Custom, OrqAdvisorTool, OrqSidekickTool, OrqAiTool, ToolsMCPTool]], # optional
"top_k": Optional[int],
"top_logprobs": Optional[int],
"top_p": Optional[float],
"variables": Dict[str, Any], # optional
}
{
background?: boolean;
cache?: {
ttl?: number;
type: string; // required
};
cacheControl?: {
ttl?: "5m" | "1h";
type: "ephemeral"; // required
};
contextManagement?: {
compactThreshold?: number;
providerNative?: boolean;
type: string; // required
}[];
conversation?: {
id: string; // required
};
fallbackOnCodes?: number[];
fallbacks?: {
model: string; // required
}[];
frequencyPenalty?: number;
guardrails?: {
executeOn: "input" | "output" | "both"; // required
id: string; // required
isGuardrail?: boolean;
sampleRate?: number;
timeout?: number;
}[];
identity?: {
displayName?: string;
email?: string;
id: string; // required
metadata?: Record<string, any>[];
tags?: string[];
};
include?: string[];
input?: string | CreateRouterResponseInput2[];
instructions?: string;
integrationId?: string;
limits?: {
maxCost?: number;
maxDepth?: number;
maxExecutionTime?: number;
maxIterations?: number;
toolTimeout?: number;
};
loadBalancer?: {
models: { // required
model: string; // required
weight: number; // required
}[];
type: "weight_based" | "round_robin" | "latency_based"; // required
};
maxOutputTokens?: number;
maxToolCalls?: number;
memory?: {
entityId: string; // required
};
metadata?: Record<string, string>;
model?: string;
parallelToolCalls?: boolean;
plugins?: {
entities?: string[];
entityThresholds?: Record<string, number>;
id: "pii_redaction" | "response_healing" | "trace_scrubbing"; // required
language?: string;
mask?: ("all" | "system" | "input" | "output" | "metadata" | "variables")[];
onFailure?: "block" | "passthrough";
regions?: string[];
threshold?: number;
}[];
presencePenalty?: number;
previousResponseId?: string;
promptCacheKey?: string;
promptCacheOptions?: {
mode?: "implicit" | "explicit";
ttl?: "30m" | "12h";
};
reasoning?: {
context?: "auto" | "current_turn" | "all_turns";
effort?: "none" | "instant" | "minimal" | "low" | "medium" | "high" | "xhigh" | "max";
summary?: "concise" | "detailed" | "auto";
};
retry?: {
count: number; // required
onCodes?: number[];
};
safetyIdentifier?: string;
security?: {
mask?: string[];
};
serviceTier?: "auto" | "default" | "flex" | "fast" | "ultrafast" | "scale" | "priority";
speed?: "standard" | "fast";
stopSequences?: string[];
store?: boolean;
stream?: boolean;
streamOptions?: {
includeObfuscation: boolean; // required
};
temperature?: number;
templateEngine?: "text" | "jinja" | "mustache";
text?: {
format?: {
// one of: PlainText, FormatJSONSchema
type: "text" | "json_schema"; // required
description?: string;
name: string; // required (variant-specific)
schema: Record<string, any>; // required (variant-specific)
strict?: boolean;
};
verbosity?: "low" | "medium" | "high";
};
thinking?: {
budgetTokens: number; // required
display?: "summarized" | "omitted" | "updates";
thinkingLevel?: "minimal" | "low" | "medium" | "high";
type: "enabled" | "disabled" | "adaptive" | "between_tools"; // required
};
thread?: {
id: string; // required
tags?: string[];
};
timeout?: {
callTimeout: number; // required
};
toolChoice?: {
// one of: Shorthand, SpecificFunction
name: string; // required (variant-specific)
type: "function"; // required (variant-specific)
};
tools?: (ToolsFunction | Custom | OrqAdvisorTool | OrqSidekickTool | OrqAiTool | MCPTool)[];
topK?: number;
topLogprobs?: number;
topP?: number;
variables?: Record<string, any>;
}
Show Response
Show Response
{
"background": bool,
"completed_at": Nullable[int],
"conversation": { # optional
"id": str,
},
"created_at": int,
"error": {
"code": str,
"message": str,
},
"frequency_penalty": float,
"id": str,
"incomplete_details": {
"reason": str,
},
"input": List[Any],
"instructions": Nullable[str],
"max_output_tokens": Nullable[int],
"max_tool_calls": Nullable[int],
"memory": { # optional
"entity_id": str,
},
"metadata": Dict[str, str],
"model": str,
"object": str,
"output": List[Any],
"parallel_tool_calls": bool,
"presence_penalty": float,
"previous_response_id": Nullable[str],
"prompt_cache_key": Nullable[str],
"prompt_cache_options": {
"mode": Optional[Literal["implicit", "explicit"]],
"ttl": Optional[Literal["30m", "12h"]],
},
"prompt_cache_retention": Nullable[str],
"reasoning": {
"context": Optional[Literal["auto", "current_turn", "all_turns"]],
"effort": Optional[Literal["none", "instant", "minimal", "low", "medium", "high", "xhigh", "max"]],
"summary": Optional[Literal["concise", "detailed", "auto"]],
},
"safety_identifier": Nullable[str],
"service_tier": Literal["auto", "default", "flex", "fast", "ultrafast", "scale", "priority"],
"status": Literal["queued", "in_progress", "completed", "failed", "incomplete"],
"store": bool,
"telemetry": { # optional
"span_id": str,
"trace_id": str,
},
"temperature": float,
"text": Any,
"tool_choice": Any,
"tools": List[Any],
"top_k": Optional[int],
"top_logprobs": int,
"top_p": float,
"truncation": Literal["disabled", "auto"],
"usage": {
"input_cost": Optional[float],
"input_tokens": int,
"input_tokens_details": {
"cache_creation_1h_tokens": Optional[int],
"cache_creation_5m_tokens": Optional[int],
"cache_creation_tokens": int,
"cache_write_tokens": int,
"cache_write_tokens_details": { # optional
"ttl_12h_tokens": int,
"ttl_30m_tokens": int,
},
"cached_tokens": int,
},
"output_cost": Optional[float],
"output_tokens": int,
"output_tokens_details": {
"reasoning_tokens": int,
},
"server_tool_use": { # optional
"advisor_requests": Optional[int],
"code_interpreter_sessions": Optional[int],
"datetime_requests": Optional[int],
"fusion_requests": Optional[int],
"image_generation_calls": Optional[int],
"search_models_requests": Optional[int],
"shell_commands": Optional[int],
"subagent_requests": Optional[int],
"tool_search_requests": Optional[int],
"web_fetch_requests": Optional[int],
"web_search_requests": Optional[int],
},
"speed": Optional[Literal["standard", "fast"]],
"total_cost": Optional[float],
"total_tokens": int,
"web_search_requests": Optional[int],
},
"user": Nullable[str],
"variables": Dict[str, Any], # optional
}
{
background: boolean;
completedAt: number;
conversation?: {
id: string;
};
createdAt: number;
error: {
code: string;
message: string;
};
frequencyPenalty: number;
id: string;
incompleteDetails: {
reason: string;
};
input: any[];
instructions: string;
maxOutputTokens: number;
maxToolCalls: number;
memory?: {
entityId: string;
};
metadata: Record<string, string>;
model: string;
object: string;
output: any[];
parallelToolCalls: boolean;
presencePenalty: number;
previousResponseId: string;
promptCacheKey: string;
promptCacheOptions: {
mode?: "implicit" | "explicit";
ttl?: "30m" | "12h";
};
promptCacheRetention: string;
reasoning: {
context?: "auto" | "current_turn" | "all_turns";
effort?: "none" | "instant" | "minimal" | "low" | "medium" | "high" | "xhigh" | "max";
summary?: "concise" | "detailed" | "auto";
};
safetyIdentifier: string;
serviceTier: "auto" | "default" | "flex" | "fast" | "ultrafast" | "scale" | "priority";
status: "queued" | "in_progress" | "completed" | "failed" | "incomplete";
store: boolean;
telemetry?: {
spanId: string;
traceId: string;
};
temperature: number;
text: any;
toolChoice: any;
tools: any[];
topK?: number;
topLogprobs: number;
topP: number;
truncation: "disabled" | "auto";
usage: {
inputCost?: number;
inputTokens: number;
inputTokensDetails: {
cacheCreation1hTokens?: number;
cacheCreation5mTokens?: number;
cacheCreationTokens: number;
cacheWriteTokens: number;
cacheWriteTokensDetails?: {
ttl12hTokens: number;
ttl30mTokens: number;
};
cachedTokens: number;
};
outputCost?: number;
outputTokens: number;
outputTokensDetails: {
reasoningTokens: number;
};
serverToolUse?: {
advisorRequests?: number;
codeInterpreterSessions?: number;
datetimeRequests?: number;
fusionRequests?: number;
imageGenerationCalls?: number;
searchModelsRequests?: number;
shellCommands?: number;
subagentRequests?: number;
toolSearchRequests?: number;
webFetchRequests?: number;
webSearchRequests?: number;
};
speed?: "standard" | "fast";
totalCost?: number;
totalTokens: number;
webSearchRequests?: number;
};
user: string;
variables?: Record<string, any>;
}
Show Variant types
Show Variant types
CreateRouterResponseInput2
{
"arguments": Optional[str],
"async_": Optional[bool],
"call_id": Optional[str],
"content": Union[str, List[CreateRouterResponseContent2]], # optional
"id": Optional[str],
"name": Optional[str],
"output": Optional[str],
"reasoning": { # optional
"effort": Optional[Literal["none", "minimal", "low", "medium", "high", "xhigh", "max"]],
},
"role": Optional[Literal["user", "assistant", "system", "developer"]],
"status": Optional[Literal["in_progress", "completed", "incomplete"]],
"type": Optional[Literal["message", "function_call", "function_call_output", "item_reference", "reasoning", "custom_tool_call", "custom_tool_call_output", "computer_call", "computer_call_output", "local_shell_call", "local_shell_call_output", "shell_call", "shell_call_output", "apply_patch_call", "apply_patch_call_output", "tool_search_call", "tool_search_output", "additional_tools", "compaction", "program", "program_output", "mcp_call", "mcp_list_tools", "mcp_approval_request", "mcp_approval_response", "configuration_update"]],
}
{
arguments?: string;
async?: boolean;
callId?: string;
content?: string | CreateRouterResponseContent2[];
id?: string;
name?: string;
output?: string;
reasoning?: {
effort?: "none" | "minimal" | "low" | "medium" | "high" | "xhigh" | "max";
};
role?: "user" | "assistant" | "system" | "developer";
status?: "in_progress" | "completed" | "incomplete";
type?: "message" | "function_call" | "function_call_output" | "item_reference" | "reasoning" | "custom_tool_call" | "custom_tool_call_output" | "computer_call" | "computer_call_output" | "local_shell_call" | "local_shell_call_output" | "shell_call" | "shell_call_output" | "apply_patch_call" | "apply_patch_call_output" | "tool_search_call" | "tool_search_output" | "additional_tools" | "compaction" | "program" | "program_output" | "mcp_call" | "mcp_list_tools" | "mcp_approval_request" | "mcp_approval_response" | "configuration_update";
}
Retrieve a Response
Retrieves a previously created response by its ID.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.responses.get(response_id="<id>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.responses.get({
responseId: "<id>",
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"response_id": str, # required
}
{
responseId: string; // required
}
Show Response
Show Response
{
"background": bool,
"completed_at": Nullable[int],
"conversation": { # optional
"id": str,
},
"created_at": int,
"error": {
"code": str,
"message": str,
},
"frequency_penalty": float,
"id": str,
"incomplete_details": {
"reason": str,
},
"input": List[Any],
"instructions": Nullable[str],
"max_output_tokens": Nullable[int],
"max_tool_calls": Nullable[int],
"memory": { # optional
"entity_id": str,
},
"metadata": Dict[str, str],
"model": str,
"object": str,
"output": List[Any],
"parallel_tool_calls": bool,
"presence_penalty": float,
"previous_response_id": Nullable[str],
"prompt_cache_key": Nullable[str],
"prompt_cache_options": {
"mode": Optional[Literal["implicit", "explicit"]],
"ttl": Optional[Literal["30m", "12h"]],
},
"prompt_cache_retention": Nullable[str],
"reasoning": {
"context": Optional[Literal["auto", "current_turn", "all_turns"]],
"effort": Optional[Literal["none", "instant", "minimal", "low", "medium", "high", "xhigh", "max"]],
"summary": Optional[Literal["concise", "detailed", "auto"]],
},
"safety_identifier": Nullable[str],
"service_tier": Literal["auto", "default", "flex", "fast", "ultrafast", "scale", "priority"],
"status": Literal["queued", "in_progress", "completed", "failed", "incomplete"],
"store": bool,
"telemetry": { # optional
"span_id": str,
"trace_id": str,
},
"temperature": float,
"text": Any,
"tool_choice": Any,
"tools": List[Any],
"top_k": Optional[int],
"top_logprobs": int,
"top_p": float,
"truncation": Literal["disabled", "auto"],
"usage": {
"input_cost": Optional[float],
"input_tokens": int,
"input_tokens_details": {
"cache_creation_1h_tokens": Optional[int],
"cache_creation_5m_tokens": Optional[int],
"cache_creation_tokens": int,
"cache_write_tokens": int,
"cache_write_tokens_details": { # optional
"ttl_12h_tokens": int,
"ttl_30m_tokens": int,
},
"cached_tokens": int,
},
"output_cost": Optional[float],
"output_tokens": int,
"output_tokens_details": {
"reasoning_tokens": int,
},
"server_tool_use": { # optional
"advisor_requests": Optional[int],
"code_interpreter_sessions": Optional[int],
"datetime_requests": Optional[int],
"fusion_requests": Optional[int],
"image_generation_calls": Optional[int],
"search_models_requests": Optional[int],
"shell_commands": Optional[int],
"subagent_requests": Optional[int],
"tool_search_requests": Optional[int],
"web_fetch_requests": Optional[int],
"web_search_requests": Optional[int],
},
"speed": Optional[Literal["standard", "fast"]],
"total_cost": Optional[float],
"total_tokens": int,
"web_search_requests": Optional[int],
},
"user": Nullable[str],
"variables": Dict[str, Any], # optional
}
{
background: boolean;
completedAt: number;
conversation?: {
id: string;
};
createdAt: number;
error: {
code: string;
message: string;
};
frequencyPenalty: number;
id: string;
incompleteDetails: {
reason: string;
};
input: any[];
instructions: string;
maxOutputTokens: number;
maxToolCalls: number;
memory?: {
entityId: string;
};
metadata: Record<string, string>;
model: string;
object: string;
output: any[];
parallelToolCalls: boolean;
presencePenalty: number;
previousResponseId: string;
promptCacheKey: string;
promptCacheOptions: {
mode?: "implicit" | "explicit";
ttl?: "30m" | "12h";
};
promptCacheRetention: string;
reasoning: {
context?: "auto" | "current_turn" | "all_turns";
effort?: "none" | "instant" | "minimal" | "low" | "medium" | "high" | "xhigh" | "max";
summary?: "concise" | "detailed" | "auto";
};
safetyIdentifier: string;
serviceTier: "auto" | "default" | "flex" | "fast" | "ultrafast" | "scale" | "priority";
status: "queued" | "in_progress" | "completed" | "failed" | "incomplete";
store: boolean;
telemetry?: {
spanId: string;
traceId: string;
};
temperature: number;
text: any;
toolChoice: any;
tools: any[];
topK?: number;
topLogprobs: number;
topP: number;
truncation: "disabled" | "auto";
usage: {
inputCost?: number;
inputTokens: number;
inputTokensDetails: {
cacheCreation1hTokens?: number;
cacheCreation5mTokens?: number;
cacheCreationTokens: number;
cacheWriteTokens: number;
cacheWriteTokensDetails?: {
ttl12hTokens: number;
ttl30mTokens: number;
};
cachedTokens: number;
};
outputCost?: number;
outputTokens: number;
outputTokensDetails: {
reasoningTokens: number;
};
serverToolUse?: {
advisorRequests?: number;
codeInterpreterSessions?: number;
datetimeRequests?: number;
fusionRequests?: number;
imageGenerationCalls?: number;
searchModelsRequests?: number;
shellCommands?: number;
subagentRequests?: number;
toolSearchRequests?: number;
webFetchRequests?: number;
webSearchRequests?: number;
};
speed?: "standard" | "fast";
totalCost?: number;
totalTokens: number;
webSearchRequests?: number;
};
user: string;
variables?: Record<string, any>;
}