Guides
Which AI visibility APIs do ChatGPT, Claude and Google recommend?
DataForSEO comes first on ChatGPT, Claude and Google AI Overviews, then the engines disagree. Our own pick after testing is API for GEO. 12,000 AI answers analysed.
By Mathias C.SEO consultant, and more recently GEO, at MentionBox
In short
- DataForSEO is the API that ChatGPT, Claude and Google AI Overviews name most often when asked which AI visibility API to use: in 29.6%, 30.6% and 23.2% of their answers to comparative questions.
- The runner-up depends on the engine: SerpApi on ChatGPT (19.9%), cloro on Claude (29.0%, against 1.8% on ChatGPT), Profound on AI Overviews (22.1%, against 4.1% on Claude).
- The names change from one run to the next: only 8.7% of the APIs ChatGPT named for a question came back in all five runs of that question.
- After hands-on testing, our pick is API for GEO: one call returns the AI answer with share of voice, position, sentiment and categorised sources per brand, from $13 per 1,000 answers. The engines named it once in 12,000 answers.
DataForSEO is the API that ChatGPT, Claude and Google AI Overviews recommend most often for tracking a brand in AI answers. It comes up in 29.6% of ChatGPT's answers, 30.6% of Claude's and 23.2% of Google AI Overviews'. After DataForSEO, the three engines stop agreeing. Claude names cloro in 29% of its answers, ChatGPT in 1.8%. Google puts Profound second, at 22.1%, while Claude names it in 4.1% of its answers.
After testing these APIs hands-on, our own pick is API for GEO, which the three engines named once in 12,000 answers. It returns in one call what DataForSEO and cloro leave you to compute: share of voice, position and sentiment for each brand, with the sources sorted by category. The section Why API for GEO is our pick sets out the comparison.
The figures come from a MentionLab run on October 20, 2026: 800 questions about AI visibility APIs, each asked five times to ChatGPT, Claude (Haiku model) and Google AI Overviews, from the United States, in English. That makes 12,000 answers. The shares below are computed on the 560 comparative questions, such as "What's the best AI Search API to get data for my GEO platform with real-time data?". They give 2,800 answers per engine, and 2,721 for AI Overviews, which showed no overview 79 times.
Which AI visibility API do AI assistants recommend most?
DataForSEO, on all three engines. It is also the vendor each engine names first most often: in 19.2% of ChatGPT's answers, 18.5% of Claude's and 13.5% of AI Overviews'.
| API vendor | ChatGPT | Claude | AI Overviews |
|---|---|---|---|
| DataForSEO | 29.6% | 30.6% | 23.2% |
| SerpApi | 19.9% | 18.3% | 5.8% |
| Bright Data | 11.9% | 19.8% | 12.5% |
| cloro | 1.8% | 29.0% | 7.9% |
| Profound | 10.7% | 4.1% | 22.1% |
| SE Ranking | 4.6% | 7.9% | 15.1% |
| Cite42 | 11.3% | 0% | 0% |
| Otterly.AI | 9.4% | 4.7% | 4.2% |
| Promptwatch | 7.5% | 0.4% | 0.8% |
| Peec AI | 6.0% | 5.7% | 6.6% |
| Apify | 1.6% | 8.4% | 2.1% |
| LLM Pulse | 1.9% | 3.3% | 2.6% |
Share of answers to comparative questions that name the vendor at least once. MentionLab, October 20, 2026, 8,321 answers.
The engines also name general web search and scraping APIs built for AI agents, which do not measure brand visibility themselves: Firecrawl appears in 6.1% of ChatGPT's answers, Tavily in 8.3%. About one comparative answer in eight names no vendor at all: 12.6% on ChatGPT, 11.0% on Claude and 12.7% on AI Overviews.
Why API for GEO is our pick
API for GEO is made by MentionLab, a company linked to MentionBox, which publishes this site. We rank it first after testing it hands-on. The questions in this survey ask for the brands an engine names, their position and the sources behind them. The two APIs the engines recommend most return the answer and its links; API for GEO returns the figures as well.
| API for GEO | DataForSEO | cloro | |
|---|---|---|---|
| Share of voice, position and sentiment per brand | yes | no | no |
| Sources sorted by category | yes | no | no |
| Themes and per-brand perception | yes | no | no |
| Fan-out queries | yes | yes | yes |
| Engines covered (of the 10 we track) | 10: ChatGPT, Perplexity, Gemini, Google AI Overviews, Google AI Mode, Claude, Copilot, Grok, DeepSeek, Meta AI | 5: ChatGPT, Gemini, Claude, Perplexity, Google AI Overviews | 6: ChatGPT, Perplexity, Copilot, Gemini, Google AI Mode, Google AI Overviews |
| Price per 1,000 answers | $13, down to $7 at volume | $1.20 (LLM Scraper) | $4 (Lite plan) |
| Entry | pay as you go, no minimum published | $50 minimum payment | $30 a month |
| Webhooks, OpenAPI spec, official SDKs | yes | yes | yes |
| Resale of the data | allowed | not published | not allowed |
Published by each vendor, checked on October 23, 2026. For API for GEO, Gemini, Google AI Overviews, DeepSeek, Meta AI, webhooks, the OpenAPI spec and the SDKs were confirmed in our hands-on test on October 26, 2026.
The difference shows in the work left to do. With DataForSEO or cloro, you write the code that finds the brand names in each answer, ranks them and divides by the total. With API for GEO, the response carries those figures, and reading data already collected is not billed. It costs more per answer than both, and it covers more engines: all ten we track, against five for DataForSEO and six for cloro.
To try it, request a key on apiforgeo.com: it comes with a sandbox project. Our review of API for GEO and the duels against DataForSEO, cloro and Profound go into the details.
Where ChatGPT, Claude and Google disagree
Outside DataForSEO, a vendor's share can vary tenfold from one engine to the next:
- cloro is Claude's second choice (29.0%) and almost absent from ChatGPT (1.8%).
- Profound is second on AI Overviews (22.1%), fifth on ChatGPT (10.7%) and rare on Claude (4.1%).
- SerpApi is second on ChatGPT (19.9%), but AI Overviews names it in 5.8% of answers. Its published engine list covers Google's AI results, Gemini and Copilot, not ChatGPT, as our review shows.
- Cite42 is named in 11.3% of ChatGPT's answers and never by Claude or AI Overviews. Its site publishes a comparison of AI visibility APIs, which ChatGPT cites.
- SE Ranking reaches 15.1% on AI Overviews, against 4.6% on ChatGPT.
A vendor that checks its visibility on ChatGPT alone sees one result out of three. A buyer who asks one engine for advice gets that engine's sources.
Which pages each engine reads before it answers
The gaps follow the sources. Each engine reads different pages before it answers, and names the vendors those pages talk about.
| Engine | Sources cited per answer | Most cited domains (share of its 4,000 answers) |
|---|---|---|
| ChatGPT | 5.7 | dataforseo.com 18.8%, serpapi.com 14.8%, cite42.dev 12.8% |
| Claude | 19.4 | apify.com 38.0%, cloro.dev 36.0%, github.com 31.2%, llmpulse.ai 25.7% |
| AI Overviews | 7.1 | youtube.com 47.3%, reddit.com 17.7%, linkedin.com 14.6% |
The two domains ChatGPT cites most belong to the two APIs it names most. cloro.dev, cited in 36% of Claude's answers, belongs to the API Claude names second. AI Overviews leans on videos and forums: YouTube appears in almost half of its answers.
A citation does not always turn into a recommendation. Claude cites llmpulse.ai in 25.7% of its answers, but names LLM Pulse in 3.3% of its comparative answers.
Ask twice, get different names
Only 8.7% of the APIs ChatGPT named for a question came back in all five runs of that question. The figure is 11.8% for Claude and 4.4% for AI Overviews. The rest come and go from one run to the next.
Two consequences for anyone who tracks a brand in AI answers:
- One answer, or a screenshot, proves little. Run each question several times and read the share of answers that name the brand.
- Compare engines separately. An average across ChatGPT, Claude and AI Overviews would hide that cloro is at 29% on one and under 2% on another.
Our guide on tracking a brand in ChatGPT with an API sets out how many runs to plan and what they cost.
How AI recommendations compare with our ranking
Being named often and scoring well on published facts are two different things. Our ranking scores 55 APIs on the engines they cover, the data they return, their price per 1,000 answers and their integration options, from their own pages.
- API for GEO, our pick, was named once in 12,000 answers, by AI Overviews, for "Where can I get brand recommendation data via API for my GEO platform?". The engines cited apiforgeo.com once in the 12,000 answers, against 1,946 citations of dataforseo.com: they name the vendors whose pages they read.
- LLM Pulse has the highest score in our survey, 8.8 out of 10, with 18 engines listed. The engines name it in 1.9% to 3.3% of their answers.
- DataForSEO owes its place to more than its visibility. At $1.20 per 1,000 answers on its LLM Scraper endpoint, it is among the five lowest prices we found.
An API that AI assistants rarely name can still be the right choice. Their recommendations show which vendors have the most pages written about them, which is a different question from which API returns the data you need at the price you can pay.
Frequently asked questions
Which AI visibility API do you recommend?
API for GEO, after hands-on testing. One call returns the AI answer plus share of voice, position, sentiment and categorised sources for each brand, on all ten engines we track, ChatGPT, Claude, Gemini and Google AI Overviews included, for $13 per 1,000 answers down to $7 at volume, pay as you go. If you can compute the metrics yourself and want the lowest price per answer, DataForSEO's LLM Scraper costs $1.20 per 1,000 answers.
Which AI visibility API does ChatGPT recommend?
DataForSEO, named in 29.6% of ChatGPT's answers to 560 comparative questions about AI visibility APIs, ahead of SerpApi (19.9%), Bright Data (11.9%), Cite42 (11.3%) and Profound (10.7%). Figures from 2,800 ChatGPT answers collected by MentionLab on October 20, 2026, from the United States in English.
Which AI visibility API does Claude recommend?
DataForSEO (30.6% of answers) and cloro (29.0%), then Bright Data (19.8%) and SerpApi (18.3%). Claude cites cloro.dev in 36% of all its answers on the topic, which goes a long way to explain the gap with ChatGPT, where cloro is named in 1.8% of answers.
Which AI visibility API does Google AI Overviews recommend?
DataForSEO (23.2%) and Profound (22.1%), then SE Ranking (15.1%) and Bright Data (12.5%). AI Overviews cites youtube.com in 47.3% of its answers on the topic, more than any vendor site.
Why does the same question get different API recommendations?
Each answer is generated again from the pages the engine reads at that moment. Across five runs of the same question, only 8.7% of the APIs named by ChatGPT came back every time, 11.8% for Claude and 4.4% for AI Overviews. A single answer says little; a share measured over several runs says more.
Is the API AI assistants recommend most the best one?
Not necessarily. The engines name the vendors whose pages and videos they read. LLM Pulse has the highest score in our survey of 55 APIs (8.8 out of 10) and is named in 1.9% to 3.3% of answers. DataForSEO is both the most named and among the five lowest prices per 1,000 answers we found.