AI Optimization API Fresh

DataForSEO AI Optimization API provides the infrastructure for AI search optimization (AIO) — the practice of improving how brands, products, and content appear in LLM-generated responses. It covers keyword discovery, conversational search analytics, LLM benchmarking, and brand mention tracking across AI platforms.

What It Covers

The AI Optimization API consists of four purpose-driven sub-APIs covering 33 total endpoints:

Sub-API	Platforms	Endpoints	What it does
LLM Responses API	ChatGPT, Claude, Gemini, Perplexity	13	Generate and retrieve structured responses from LLMs based on your input parameters
LLM Scraper API	ChatGPT Search, Gemini Search	12	Retrieve actual scraped AI search results — what users see in the AI search interface
AI Keyword Data API	All platforms	2	Search volume estimates and user intent insights based on keyword usage across AI tools
LLM Mentions API	All platforms	6	Track how often a brand, domain, or keyword is mentioned in LLM responses

Use Cases

• Brand visibility monitoring — Measure how often your brand appears when users ask AI tools about your category
• Competitor benchmarking — Compare your AI mention count against competitors across platforms
• Content optimization for AI — Identify what topics and phrases drive LLM citations for your domain
• AI keyword research — Discover which terms your target audience uses in AI-powered searches
• Conversational query testing — Test how different LLMs respond to queries about your products or services

All Endpoints (33 total)

LLM Mentions (6 endpoints)

Endpoint	Method	Description
/v3/ai_optimization/llm_mentions/search/live	POST	Search for brand/keyword mentions across all LLM platforms
/v3/ai_optimization/llm_mentions/live	POST	Get mention overview and stats
/v3/ai_optimization/llm_mentions/tasks_post	POST	Queue async mention tasks
/v3/ai_optimization/llm_mentions/tasks_ready	GET	Get completed mention tasks
/v3/ai_optimization/llm_mentions/tasks_fixed	GET	Get re-parsed mention tasks
/v3/ai_optimization/llm_mentions/task_get/{id}	GET	Retrieve mention task results

ChatGPT LLM Responses (4 endpoints)

Endpoint	Method
/v3/ai_optimization/chatgpt/tasks_post	POST
/v3/ai_optimization/chatgpt/tasks_ready	GET
/v3/ai_optimization/chatgpt/tasks_fixed	GET
/v3/ai_optimization/chatgpt/task_get/{id}	GET

ChatGPT LLM Scraper (6 endpoints)

Endpoint	Method
/v3/ai_optimization/chatgpt/llm_scraper/task_post	POST
/v3/ai_optimization/chatgpt/llm_scraper/tasks_ready	GET
/v3/ai_optimization/chatgpt/llm_scraper/tasks_fixed	GET
/v3/ai_optimization/chatgpt/llm_scraper/task_get/{id}	GET
/v3/ai_optimization/chatgpt/llm_scraper/live	POST
/v3/ai_optimization/chatgpt/llm_scraper/locations	GET

Claude LLM Responses (4 endpoints)

See Claude Responses — Standard method only.

Endpoint	Method
/v3/ai_optimization/claude/tasks_post	POST
/v3/ai_optimization/claude/tasks_ready	GET
/v3/ai_optimization/claude/tasks_fixed	GET
/v3/ai_optimization/claude/task_get/{id}	GET

Gemini LLM Responses (4 endpoints) + LLM Scraper (6 endpoints)

See Gemini Responses

Perplexity LLM Responses (1 endpoint)

See Perplexity Responses

Endpoint	Method
/v3/ai_optimization/perplexity/live	POST

AI Keyword Data (2 endpoints)

See AI Keyword Data

Endpoint	Method
/v3/ai_optimization/ai_keyword_data/live	POST
/v3/ai_optimization/ai_keyword_data/locations	GET

---

Supported Platforms

Platform	LLM Responses	LLM Scraper	LLM Mentions
ChatGPT (OpenAI)	Yes (Standard)	Yes (Live + Standard)	Yes
Claude (Anthropic)	Yes (Standard)	No	Yes
Gemini (Google)	Yes (Standard)	Yes (Live + Standard)	Yes
Perplexity	Yes (Live only)	No	Yes

Methods

The AI Optimization API supports two retrieval methods depending on the sub-API:

Live method — Instant results in a single POST request. No separate GET request needed. Used by LLM Mentions API and AI Keyword Data API. Also available for LLM Responses on supported platforms.

Standard method — Asynchronous: POST to set task, GET to retrieve results. More affordable than Live. Supports pingback_url and postback_url for webhook-style delivery. Used by LLM Responses and LLM Scraper APIs on platforms where Live is not available.

Sub-API	Live	Standard
LLM Responses	Yes (platform-dependent)	Yes
LLM Scraper	No	Yes
AI Keyword Data	Yes	No
LLM Mentions	Yes	No

Execution Times

• Live method: Up to 120 seconds per request (LLM response generation takes time)
• Standard method: May take up to 72 hours to complete

Rate Limits

• Up to 2,000 API calls per minute
• Maximum 30 simultaneous Live requests per account per platform

Quick Authentication Example

curl --request POST \
  --url https://api.dataforseo.com/v3/ai_optimization/llm_mentions/search/live \
  --header 'Authorization: Basic BASE64(login:password)' \
  --header 'Content-Type: application/json' \
  --data '[{"keyword": "your brand name", "location_code": 2840, "language_code": "en"}]'

Python Setup

import requests
from requests.auth import HTTPBasicAuth
import json

LOGIN = "your_login"
PASSWORD = "your_password"

def ai_optimization_request(path, payload):
    url = f"https://api.dataforseo.com/v3/ai_optimization/{path}"
    response = requests.post(
        url,
        auth=HTTPBasicAuth(LOGIN, PASSWORD),
        headers={"Content-Type": "application/json"},
        data=json.dumps(payload)
    )
    return response.json()

Sandbox Testing

DataForSEO Sandbox is available for free testing. Sandbox requests return mock data in the same schema as production and do not consume credits.

Pricing

Cost varies by sub-API and method. Live method requests are priced higher than Standard. Check your account dashboard for current per-request pricing and spending summaries.