Get Agent Peak RPM

Retrieve the peak RPM (requests per minute) of a specific Agent over a time period, while also returning the time when the peak occurred, the average RPM, and the total number of requests.

Note: You can query up to 31 days of data; statistics only cover developer space channels. For the scope of statistics, see "Statistical Caliber" below.

Request Method

GET

Endpoint

https://api-${endpoint}.gptbots.ai/v1/org/agent/stat/rpm

Authentication

Use account-level DevKey / DevSecret for Basic authentication: Authorization: Basic ${BASIC_TOKEN}, where ${BASIC_TOKEN} is the Base64 encoding of DevKey:DevSecret.

The calling account must be a member of the organization corresponding to org_id and have "View Usage" permission (organization Owners and administrators have this by default) and must be able to view this Agent: the Agent creator, organization Owner, or a member granted permission for this Agent (or all Agents).
Please call the endpoint of the data center where the organization is located. ${endpoint} is obtained from the region returned by Query Organization List (SG → sg, TH → th, JP → jp). Calling a different data center will return "Not a member of this organization."

Call Frequency

Each developer key can call this API up to 60 times per minute.
Each call counts toward the organization's package "Statistical API" per-minute quota, shared with Agent API Key calls for statistical interfaces.
Data is updated hourly; repeated queries within the same hour return the same result, so it is recommended to call the API once per hour.

Request

Example Request

curl -X GET 'https://api-${endpoint}.gptbots.ai/v1/org/agent/stat/rpm?org_id=${org_id}&agent_id=${agent_id}&start_time=1788192000000&end_time=1788796800000' \ -H 'Authorization: Basic ${BASIC_TOKEN}'

                      
                      curl -X GET 'https://api-${endpoint}.gptbots.ai/v1/org/agent/stat/rpm?org_id=${org_id}&agent_id=${agent_id}&start_time=1788192000000&end_time=1788796800000' \
  -H 'Authorization: Basic ${BASIC_TOKEN}'

This code block in the floating window

Request Header

Field	Type	Description
Authorization	Basic ${BASIC_TOKEN}	Basic Token after `DevKey:DevSecret` Base64 encoding.

Request Parameters

Field	Type	Required	Description
org_id	String	Yes	Organization ID.
agent_id	String	Yes	Agent ID. Workflow is not supported.
start_time	Long	Yes	Start time, in milliseconds timestamp. Rounded down to the nearest hour.
end_time	Long	Yes	End time (exclusive), in milliseconds timestamp. Rounded up to the nearest hour. The aligned interval cannot exceed 31 days.
conversation_type	String	No	View only a specific channel, such as `API`, `SHARE`, `EMBED`, `WIDGET`, `LIVEDESK`. Can only be a channel from the development space; if not provided, all channels in the development space are counted.

Response

Response Example

{ "code": 0, "message": "OK", "data": { "agent_id": "65f0c2a1b4e8d93a1c7e5b21", "range_start": 1788192000000, "range_end": 1788796800000, "data_until": 1788793200000, "peak_rpm": 42, "peak_at": 1788512460000, "avg_rpm": 3.18, "total_requests": 18234 } }

                      
                      {
    "code": 0,
    "message": "OK",
    "data": {
        "agent_id": "65f0c2a1b4e8d93a1c7e5b21",
        "range_start": 1788192000000,
        "range_end": 1788796800000,
        "data_until": 1788793200000,
        "peak_rpm": 42,
        "peak_at": 1788512460000,
        "avg_rpm": 3.18,
        "total_requests": 18234
    }
}

This code block in the floating window

Successful Response

Field	Type	Description
agent_id	String	Agent ID.
range_start	Long	Start of the actual statistics interval: `start_time`, aligned down to the nearest whole hour in milliseconds timestamp.
range_end	Long	End of the actual statistics interval (exclusive): `end_time`, aligned up to the nearest whole hour in milliseconds timestamp.
data_until	Long	The time until which data has been counted (milliseconds timestamp). Parts of the interval after this time do not yet have data.
peak_rpm	Long	Peak RPM: the number of requests in the minute with the highest total requests across all channels within the interval.
peak_at	Long	Start of the minute with the peak (milliseconds timestamp). If multiple minutes tie, take the earliest; if there are no requests in the interval, `null`.
avg_rpm	Number	Average RPM: total requests ÷ number of minutes with requests, rounded to 2 decimal places.
total_requests	Long	Total number of requests in the interval.

Failed Response

Field	Type	Description
code	Integer	Error code.
message	String	Error details.

Error Codes

HTTP status code corresponds to the first three digits of the error code. For example, 40000 corresponds to 400, 40100 corresponds to 401, 403132 corresponds to 403.

Error Code	Description
40000	Parameter error, such as missing required parameters or invalid parameter values.
40001	Too many requests: this developer key has exceeded the per-minute limit for this interface.
40008	Exceeded the organization's package "statistical API" per-minute quota.
40010	Interface temporarily closed, please try again later.
40100	No permission: Missing usage viewing permission, or no right to view this Agent (including the Agent not belonging to this organization).
40105	Not a member of this organization, or accessed an address from another data center.
40324	DevKey or DevSecret error.
40378	Agent has been deleted.
403132	Invalid time range: start time is not earlier than end time, or exceeds the maximum query range.

Statistical Scope

Requests: User-submitted question messages, plus dialogue errors (including requests rejected due to exceeding call frequency), consistent with "Agent Insights → Requests per Minute" in the console. It is not the number of HTTP requests, and may differ from the counting in package rate limits.
Statistical Range: Only counts channels in the development space, including API, shared pages, web embedding, bubble components, LiveDesk, and various instant messaging and phone channels; does not include debugging previews in the development space, Workflow executions, or usage of the Agent in a workspace after deployment in that workspace.
Update Frequency: Data is collected on the hour every hour, with about a 1-hour delay, based on data_until. Data before 2026-03-09 has no minute-level details, and query results are 0.