DataForSEO
SEO, search-engine results, keyword research, backlinks and AI visibility data.
Live LLM Mentions
The individual answers in which an assistant mentioned your target. Returns total_count, current_offset, search_after_token and items - page with the token, not offset, past the first pages. ⚠️ target is an array of objects, each {"domain": "..."} or {"keyword": "..."}; a bare string is rejected as the wrong type and an array of strings as 'Each target item must be an object'. Filter fields come from get_dataforseo_ai_llm_mentions_available_filters. Wrapped in DataForSEO's envelope: data in tasks[0].result, outcome in tasks[0].status_code - a rejected request still returns HTTP 200. For counts rather than the mentions themselves use post_dataforseo_ai_llm_mentions_aggregated_metrics_live.
- 型: array object[]
- target型: array object[]必須
array of objects containing target entities required field you can specify up to 10 entities (objects) in the target field one target entity can contain either one domain or one keyword and related parametersexamples: target array with a domain entity [{"domain": "en.wikipedia.org", "search_filter": "exclude"}] target array with a keyword entity [{"keyword": "bmw", "search_scope": ["question"], "match_type ": "partial_match"}] target array with multiple entities [{"domain": "en.wikipedia.org", "search_filter": "exclude"}, {"keyword": "bmw", "match_type ": "partial_match", "search_scope": ["answer"]}]
- domain型: string必須
target domain required field if you don't specify keyword you can specify up to 63 characters in the domain field; a domain should be specified without https:// and www.
- include型: boolean
_subdomains indicates if the subdomains of the target domain will be included in the search optional field if set to true, the subdomains will be included in the search default value: false
- keyword型: string
target keyword required field if you don't specify domain you can specify up to 250 characters in the keyword field all %## will be decoded (plus character ‘+’ will be decoded to a space character) if you need to use the “%” character for your keyword, please specify it as “%25”; if you need to use the “+” character for your keyword, please specify it as “%2B”learn more about rules and limitations of keyword and keywords fields in DataForSEO APIs in this Help Center article
- match型: string
_type target keyword match type defines how the specified keyword is matched optional field possible values: word_match - full-text search for terms that match the specified seed keyword with additional words included before, after, or within the key phrase (e.g., search for "light" will return results with "light bulb", "light switch"); partial_match - substring search that finds all instances containing the specified sequence of characters, even if it appears inside a longer word (e.g., search for "light" will return results with "lighting", "highlight"); default value: word_match
- search型: string
_filter Entity search filter: include or exclude; defaults to include.
- search型: array string[]
_scope Search scope. Domain entities: any, sources, search_results. Keyword entities: any, question, answer, brand_entities, fan_out_queries. Default: any.
- filters型: array
array of results filtering parameters optional field you can add several filters at once (8 filters maximum) you should set a logical operator and, or between the conditions the following operators are supported: =, , in, not_in, like, not_like, ilike, not_ilike, match, not_match you can use the % operator with like and not_like to match any string of zero or more characters example: ["ai_search_volume",">","1000"]The full list of possible filters is available here.
- language型: string
_code search language code optional field if you use this field, you don't need to specify language_name; you can receive the list of available languages of the search engine with their language_code_by making a separate request to the https://api.dataforseo.com/v3/ai_optimization/llm_mentions/locations_and_languages default value: en Note: chat_gpt data is available for en onlyn
- language型: string
_name full name of search language optional field if you use this field, you don't need to specify language_code; if you don't specify this field, the language_code with en value will be used by default; you can receive the list of available languages of the search engine with their language_name by making a separate request to the https://api.dataforseo.com/v3/ai_optimization/llm_mentions/locations_and_languages Note: chat_gpt data is available for English only
- limit型: integer
the maximum number of returned objects optional fielddefault value: 100 maximum value: 1000
- location型: integer
_code search location code optional field if you use this field, you don't need to specify location_name you can receive the list of available locations of the search engine with their location_code by making a separate request to the https://api.dataforseo.com/v3/ai_optimization/llm_mentions/locations_and_languages default value: 2840 Note: chat_gpt data is available for 2840 only
- location型: string
_name full name of search location optional field if you use this field, you don't need to specify location_code if you don't specify this field, the location_code with 2840 value will be used by default; you can receive the list of available locations of the search engine with their location_name by making a separate request to the https://api.dataforseo.com/v3/ai_optimization/llm_mentions/locations_and_languages Note: chat_gpt data is available for United States only
- offset型: integer
offset in the results array of the returned mentions data optional fielddefault value: 0 example: if you specify the 10 value, the first ten mentions objects in the results array will be omitted and the data will be provided for the successive objects; Note: the maximum value is 9,000, use the search_after_token if you would like to offset more results
- order型: array string[]
_by results sorting rules optional field you can use the same values as in the filters array to sort the results possible sorting types: asc - results will be sorted in the ascending order desc - results will be sorted in the descending order you should use a comma to set up a sorting type example: ["ai_search_volume,desc"] note that you can set no more than three sorting rules in a single request you should use a comma to separate several sorting rules
- platform型: string
target platform optional field possible values: chat_gpt, google default value: google Note: the data returned depends on the selected platform Note #2:chat_gpt data is available for the United States and English only
- search型: string
_after _token token for subsequent requests optional field provided in the identical filed of the response to each request; use this parameter to avoid timeouts while trying to obtain over 20,000 results in a single request; by specifying the unique search_after_token value from the response array, you will get the subsequent results of the initial task; search_after_token values are unique for each subsequent task ; Note: if the search_after_token is specified in the request, all other parameters should be identical to the previous request
- tag型: string
user-defined task identifier optional field the character limit is 255 you can use this parameter to identify the task and match it with the result you will find the specified tag value in the data object of the response
- 型: object200
Successful response
- 型: numbercost
total tasks cost, USD
- 型: integerstatus
_code general status code you can find the full list of the response codes here Note: we strongly recommend designing a necessary system for handling related exceptional or error conditions
- 型: stringstatus
_message general informational message you can find the full list of general informational messages here
- 型: array string[]tasks
array of tasks
- 型: integertasks
_count the number of tasks in the tasks array
- 型: integertasks
_error the number of tasks in the tasks array returned with an error
- 型: numbertasks
.cost cost of the task, USD
- 型: objecttasks
.data contains the same parameters that you specified in the POST request
- 型: stringtasks
.id task identifier unique task identifier in our system in the UUID format
- 型: array string[]tasks
.path URL path
- 型: array string[]tasks
.result array of results
- 型: integertasks
.result _count number of elements in the result array
- 型: integertasks
.result .current _offset the number of mentions objects that are omitted in the items array
- 型: array string[]tasks
.result .items contains relevant mentions data
- 型: integertasks
.result .items _count the number of results returned in the items array
- 型: integertasks
.result .items .ai _search _volume current AI search volume rate of a keyword learn more about this metric here
- 型: stringtasks
.result .items .answer relevant answer in markdown format content of the result formatted in the markdown markup language
- 型: array string[]tasks
.result .items .brand _entities array of brand entities contains information on brands mentioned in the response
- 型: stringtasks
.result .items .brand _entities .category category of the brand
- 型: integertasks
.result .items .brand _entities .position position in the results
- 型: stringtasks
.result .items .brand _entities .title name of the brand
- 型: array string[]tasks
.result .items .fan _out _queries array of fan-out queries contains related search queries derived from the main query to provide a more comprehensive response
- 型: stringtasks
.result .items .first _response _at date and time when the response data was first recorded in the UTC format: “yyyy-mm-dd hh-mm-ss +00:00” example: 2025-10-21 06:25:30 +00:00
- 型: stringtasks
.result .items .language _code language code in a POST array
- 型: stringtasks
.result .items .last _response _at date and time when the response data was last updated in the UTC format: “yyyy-mm-dd hh-mm-ss +00:00” example: 2025-10-21 06:25:30 +00:00
- 型: integertasks
.result .items .location _code location code in a POST array
- 型: stringtasks
.result .items .model _name name of the AI model from which the data was retrieved Note: for the google platform type, the value is always google_ai_overview
- 型: array string[]tasks
.result .items .monthly _searches monthly AI search volume rates array of objects with AI search volume rates in a certain month of a year
- 型: integertasks
.result .items .monthly _searches .month month
- 型: integertasks
.result .items .monthly _searches .search _volume AI search volume rate in a certain month of a year learn more about this metric here
- 型: integertasks
.result .items .monthly _searches .year year
- 型: stringtasks
.result .items .platform platform received in a POST array
- 型: stringtasks
.result .items .question relevant question
- 型: array string[]tasks
.result .items .search _results array of search results all web search outputs the model retrieved when looking up information, including duplicates and unused entries
- 型: string
breadcrumb
- 型: stringtasks
.result .items .search _results .description result description
- 型: stringtasks
.result .items .search _results .domain result domain
- 型: integertasks
.result .items .search _results .position position in the results
- 型: stringtasks
.result .items .search _results .publication _date date and time when the result was published in the format: “year-month-date:minutes:UTC_difference_hours:UTC_difference_minutes” example: 2019-11-15 12:57:46 +00:00
- 型: stringtasks
.result .items .search _results .title result title
- 型: stringtasks
.result .items .search _results .url result URL
- 型: array string[]tasks
.result .items .sources array of sources the sources the model cited or relied on in its final answer
- 型: stringtasks
.result .items .sources .domain source domain
- 型: stringtasks
.result .items .sources .markdown content of the element in markdown format content of the result formatted in the markdown markup language
- 型: integertasks
.result .items .sources .position position in the results
- 型: stringtasks
.result .items .sources .publication _date date and time when the result was published in the format: “year-month-date:minutes:UTC_difference_hours:UTC_difference_minutes” example: 2019-11-15 12:57:46 +00:00
- 型: stringtasks
.result .items .sources .snippet source description
- 型: stringtasks
.result .items .sources .source _name source name
- 型: stringtasks
.result .items .sources .thumbnail source thumbnail
- 型: stringtasks
.result .items .sources .title source title
- 型: stringtasks
.result .items .sources .url source URL
- 型: stringtasks
.result .search _after _token token for subsequent requests by specifying the unique search_after_token when setting a new task, you will get the subsequent results of the initial task; search_after_token values are unique for each subsequent task
- 型: integertasks
.result .total _count total amount of results relevant the request
- 型: integertasks
.status _code status code of the task generated by DataForSEO; can be within the following range: 10000-60000 you can find the full list of the response codes here
- 型: stringtasks
.status _message informational message of the task you can find the full list of general informational messages here
- 型: stringtasks
.time execution time, seconds
- 型: stringtime
execution time, seconds
- 型: stringversion
the current version of the API
application/json - 400
Bad request
- 401
Unauthorized
- 402
The request cannot proceed because a billing requirement is not met.
- 403
The account is not permitted to perform this operation.
- 429
Rate limit exceeded
- 500
Internal server error
- 502
The service could not complete the request.
- 503
The API or billing service is temporarily unavailable.
- 504
The service timed out while processing the request.
- default
The operation failed. Keep the response request ID when contacting Felo support.
curl https://openapi.felo.ai/v1/beta/dataforseo/ai_optimization/llm_mentions/search/live \
--request POST \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer YOUR_SECRET_TOKEN' \
--data '[
{
"target": [
{}
],
"location_name": "",
"location_code": 1,
"language_name": "",
"language_code": "",
"platform": "",
"filters": [],
"order_by": [
""
],
"offset": 1,
"search_after_token": "",
"limit": 1,
"tag": ""
}
]'
{
"version": "string",
"status_code": 1,
"status_message": "string",
"time": "string",
"cost": 1,
"tasks_count": 1,
"tasks_error": 1,
"tasks": [
"string"
],
"tasks.id": "string",
"tasks.status_code": 1,
"tasks.status_message": "string",
"tasks.time": "string",
"tasks.cost": 1,
"tasks.result_count": 1,
"tasks.path": [
"string"
],
"tasks.data": {},
"tasks.result": [
"string"
],
"tasks.result.total_count": 1,
"tasks.result.current_offset": 1,
"tasks.result.search_after_token": "string",
"tasks.result.items_count": 1,
"tasks.result.items": [
"string"
],
"tasks.result.items.platform": "string",
"tasks.result.items.model_name": "string",
"tasks.result.items.location_code": 1,
"tasks.result.items.language_code": "string",
"tasks.result.items.question": "string",
"tasks.result.items.answer": "string",
"tasks.result.items.sources": [
"string"
],
"tasks.result.items.sources.snippet": "string",
"tasks.result.items.sources.source_name": "string",
"tasks.result.items.sources.thumbnail": "string",
"tasks.result.items.sources.markdown": "string",
"tasks.result.items.sources.position": 1,
"tasks.result.items.sources.title": "string",
"tasks.result.items.sources.domain": "string",
"tasks.result.items.sources.url": "string",
"tasks.result.items.sources.publication_date": "string",
"tasks.result.items.search_results": [
"string"
],
"tasks.result.items.search_results.description": "string",
"tasks.result.items.search_results.breadcrumb": "string",
"tasks.result.items.search_results.position": 1,
"tasks.result.items.search_results.title": "string",
"tasks.result.items.search_results.domain": "string",
"tasks.result.items.search_results.url": "string",
"tasks.result.items.search_results.publication_date": "string",
"tasks.result.items.ai_search_volume": 1,
"tasks.result.items.monthly_searches": [
"string"
],
"tasks.result.items.monthly_searches.year": 1,
"tasks.result.items.monthly_searches.month": 1,
"tasks.result.items.monthly_searches.search_volume": 1,
"tasks.result.items.first_response_at": "string",
"tasks.result.items.last_response_at": "string",
"tasks.result.items.brand_entities": [
"string"
],
"tasks.result.items.brand_entities.position": 1,
"tasks.result.items.brand_entities.title": "string",
"tasks.result.items.brand_entities.category": "string",
"tasks.result.items.fan_out_queries": [
"string"
]
}