Free AI Rank Tracker in 2026: Free Tiers, Tools and an API

Search “free AI rank tracker” and you will find plenty of buttons that say free. Most of them run one check and stop. A few are a seven-day trial with a card form after it.

We read each vendor’s pricing and docs pages on 2026-10-23 and did the maths from them. Nothing here comes from running anyone’s product, so there are no speed or accuracy numbers. Where a page printed no figure we write “not verified.” Serpent sells one of the options below, so every sum is shown. For the record, its API gives 10 free calls, then $20 per 1,000 single-engine checks at Default, or $2 per 1,000 after a $100 deposit.

The options at a glance (read 2026-10-23)

One row per option. “Free part” is what you get before paying. Units differ, so read the second column before the fourth. A dash means the page printed nothing we could use.

What is genuinely free: one-off checkers and trials

Four vendors run a free check with no payment: HubSpot, Ahrefs, Trakkr and FixAEO. They answer one question, “what does an AI say about my brand right now?” They are useful for a first look. None of them is a tracker.

HubSpot AI Search Grader

HubSpot’s grader page says “there is no cost to run an AI Search Grader analysis” and that no account is needed. It reports “what ChatGPT, Perplexity, and Gemini say about you based on their training data” as a score out of 100. You fill a form with company, location, product and industry, and every field is required.

It is a one-time snapshot. HubSpot sells ongoing monitoring separately as HubSpot AEO, listed at $50 a month.

Ahrefs, Trakkr and FixAEO

Ahrefs Brand Radar lists an AI Visibility Checker under free tools that previews your visibility. Custom Prompts start at $50 a month, and the AI Visibility Index starts at $199 a month with 2,500+ checks. The page does not say Brand Radar has an API.

Trakkr’s free tool needs no login. Its page says it asks an OpenAI model the questions your buyers ask “with no web search,” and caches results for about 24 hours. Paid plans start at $100 a month for one brand and 50 prompts, with a 14-day trial.

FixAEO offers a free one-off check on any URL. Its page says the free check runs on Gemini only. Paid plans are $25 to $29 a month (Lite) and $79 (Growth).

Trials are clocks, not free tiers

Most dashboards start with a trial. Profound’s trial is free: 50 prompts, run daily for 7 days, on three engines (ChatGPT, Gemini and Google AI Overviews), with no API, no data history and no export. Scrunch, Rankscale and Semrush each list a 7-day trial, and Writesonic and Otterly list a free trial without a length we could confirm.

That is long enough to see a dashboard, not long enough to see a trend. A trend needs weeks of repeat checks on the same prompts.

What building your own tracker involves

Whatever data source you pick, a rank tracker is more than a block of answer text. Here is the work, in plain terms.

The pieces you build

• A prompt list. The questions your buyers ask, by country and language, kept in one place.
• Scheduling and storage. Rank is a trend. You need a database, a scheduler and a way to re-run failed checks.
• Citation parsing. The answer is prose. Sources arrive as separate structures, so you extract the cited links yourself.
• Normalise and match. Strip tracking parameters, handle www. and subdomains, then find your domain and its position.
• Scoring. Decide how a position becomes a number, and keep the formula fixed so charts stay comparable.
• Repeat runs. Answers vary, so one check proves little. You need the same prompt, several times.

The same prompt does not always give the same answer

AI answers vary from run to run, and between accounts and locations. Our earlier repeatability study, published in August and not a new test for this post, ran 10 prompts twice on the same day across four AI Rank engines. Citation counts changed in 15 of 40 engine pairs. One snapshot over-reads the data, so track trends over repeated runs.

Different tools measure different things

Trakkr says its free check uses an OpenAI model with no web search. HubSpot says its grader reports what models say based on training data. Those are two different methods, and neither is a daily tracker.

The practical rule is simple. Compare numbers only from the same method, and write the method down next to every chart.

Paid dashboards, and what they cost

If you want charts, alerts and competitor views without writing code, a dashboard is the right product. Here are the entry points we could read, and what each includes.

• Otterly.AI. Lite is $29 a month for 15 prompts on four engines (ChatGPT, Google AI Overviews, Perplexity, Microsoft Copilot). Standard is $189 for 100 prompts and lists 2,000 API requests a month. The page shows a second, lower price set under a 15% annual toggle, without labelling it.
• Semrush. The AI Visibility Toolkit Base plan is $99 a month per domain, billed annually, for 25 custom prompts with daily rankings. The API is not in Base.
• Writesonic. $79 a month billed annually for 50 prompts on three engines. Its page does not mention an API.
• Rankscale. Pro is $99 for 1,200 credits and up to 4,800 answers. REST API access starts at Pro+ ($229, up to 12,000 answers). Dividing price by answers gives about $20.63 per 1,000 on Pro and $19.08 on Pro+ (our arithmetic, at the upper end of “up to”).
• Scrunch. $250 a month billed annually ($300 monthly) for 350 custom prompts. Its data API is on Enterprise only.
• Profound and Peec AI. Profound prints no dollar price. Peec printed no amounts to us, and its help page says the Starter plan has no API. Both prices are not verified.

For a side-by-side of these tools on features, see our AI visibility tools comparison and the audit of 14 AI rank trackers. The short version: these are good products, they sell seats and prompts, and API access usually sits on a higher plan.

The near-free AI rank APIs

If you want JSON you can put in your own report, dashboard or agent, three routes are left: Serpent, DataForSEO and SE Ranking’s API.

1. Serpent API

Serpent sells AI rank checks as an API with no monthly plan. You call GET /api/ai/rank/<engine> for one engine or GET /api/ai/rank for several. The four engines are ChatGPT, Claude, Gemini and Perplexity, and each response carries a model field with that engine’s public label: gpt-5.5, claude-sonnet-4-6, gemini-3.1-pro and sonar-pro.

What one call returns

• The answer text in response_text.
• Cited sources: position, url, url_normalized, domain, title and cited_text. The landing page lists up to 10 citations per engine, or 15 with prompt_type=deep.
• When you pass domain: target_found, target_position, target_match_type (exact or subdomain) and target_matched_domain.
• A run summary: visibility_score (0 to 100, or null with no domain), found_in, best_position and all_domains_cited.
• Options: prompt_type (standard, deep or brand), country, language and engines.

What it costs

Rates are per call and flat. Citation count and prompt style do not change them.

Two details matter. A call naming two or more engines bills at the combined rate, so two engines cost the same as four. Ask for all four to get the saving. And new credit uses the schedule current when you deposit, so a $100 deposit prices that credit at Growth rates.

Where Serpent is not the right call

• It is an API, not a dashboard. You store results, draw charts and send alerts yourself. Our Python tracker guide shows one way to do it.
• At Default it is not the lowest. $20 per 1,000 single-engine checks is above DataForSEO’s listed per-page rates. Growth and Scale change the picture.
• The score is our own formula. It weights engines at 35, 25, 25 and 15, so a single-engine call tops out at that engine’s share. Compare scores only across runs that used the same engines.
• Answers take time. The docs advise a client timeout of at least 120 seconds.
• The free pool is small. A free account gets 10 calls shared across services.

The pricing page has the full table, and the AI Rank API page compares the engines one by one.

2. DataForSEO LLM Scraper

DataForSEO’s LLM Scraper pricing page bills per results page: $0.0012 on the standard queue (up to 45 minutes), $0.0024 on priority (up to 5 minutes) and $0.004 live (about 120 seconds on average). That is $1.20, $2.40 and $4.00 per 1,000. The page states “To date, only ChatGPT and Gemini are supported,” so there is no Claude or Perplexity. It did not print free credits or a minimum deposit.

On the standard queue it is cheaper per page than Serpent at Default or Growth. It is a fair pick if you only need ChatGPT and Gemini and can wait.

3. SE Ranking API

SE Ranking’s API page lists an AI Search API among its standalone options, from $50 for 250,000 credits pay as you go with a $50 minimum deposit, and a no-card free trial. The page does not print how many credits an AI call uses, so we cannot give a per-check price.

What free really costs you

You get a photo, not a film

A one-off check shows today. It cannot say whether you moved. Profound’s trial lists no data history and no export, and HubSpot’s grader is a one-time snapshot by its own description. Trackers earn their keep over weeks.

A narrow slice of the engines

Free checks run on one model (Trakkr, FixAEO) or three (HubSpot, Profound trial). If your buyers use Claude or Gemini, a ChatGPT-only free check is blind to them.

Time, then clocks, then forms

The free API route costs engineering time: extraction, matching, storage and re-running failed checks. Trials end after 7 or 14 days. Gemini’s free tier carries limits Google says are not guaranteed. And checkers often want a form filled, which HubSpot’s does.

Settings that quietly change the answer

Country, language and prompt wording all shape a result. If you do not record them, a change in your chart has no explanation. Store the model label and your parameters with every row.

Cost at 1K, 10K, 100K and 1M checks

This is arithmetic from the published rates above, not a measurement. The first six columns are Serpent’s own tiers. DataForSEO is per results page, one engine.

How to read it

• The deposit tiers are the whole story. Default to Growth cuts the rate by 90%. A $100 deposit buys 25,000 four-engine checks at Growth rates ($100 divided by $0.004).
• Four engines is not four times one. At Default, four engines at $40 per 1,000 is $10 an engine, against $20 for one.
• DataForSEO is cheaper per page than Serpent at Default and Growth, and Serpent at Scale ($1.00) is below its $1.20 standard queue. DataForSEO covers two engines, not four.
• Building your own is not free. The table counts API fees only, before the storage, scheduling and parsing code described above.

Per-prompt plans, compared fairly

Dashboards price by prompts, so convert. Tracking 50 prompts once a day for 30 days is 1,500 four-engine checks. On Serpent that is $60 at Default, $6 at Growth and $3 at Scale. For 15 prompts it is 450 checks: $18, $1.80 and $0.90.

Compare with Otterly’s $29 a month for 15 prompts or Writesonic’s $79 for 50. At Default, Serpent’s $18 for 15 prompts is below Otterly’s $29, and its $60 for 50 prompts is below Writesonic’s $79. That is API fees only, and you build the dashboard. After a $100 deposit the API is far cheaper. Vendors also count prompts and checks differently, and none of their pages told us how often each prompt runs, so treat this as a rough guide.

How to choose

• You want a first look, today: use a free checker (HubSpot, Ahrefs, Trakkr or FixAEO) and treat it as a snapshot.
• You want charts and alerts without code: start a dashboard trial and compare prompt limits before the clock ends.
• You want JSON in your own report, agent or client dashboard: use an API. For four engines and a domain-match field in one shape, Serpent is the pick from the options we read. For ChatGPT and Gemini only, on a queue, check DataForSEO too.
• You only need a few Gemini calls: Google’s free tier may do, as long as you accept limits that are not guaranteed.
• You want to compare tools: read our comparison of AI visibility tools and build-your-own tracker guide.

If you track rankings in classic search too, see our guide to a free rank tracking API.

Quick start

A single request is enough to start. This is the call from the API docs, for all four engines at once:

curl -G "https://apiserpent.com/api/ai/rank" \
  --data-urlencode "q=best project management tools" \
  -d "domain=example.com" \
  -H "X-API-Key: YOUR_API_KEY"

And one engine on its own, in Python. Set a long timeout, because answers take longer than search pages:

import os
import requests

resp = requests.get(
    "https://apiserpent.com/api/ai/rank/chatgpt",
    params={
        "q": "best project management tools",
        "domain": "example.com",
        "prompt_type": "deep",
        "country": "us",
    },
    headers={"X-API-Key": os.environ["SERPENT_API_KEY"]},
    timeout=120,
)
row = resp.json()["results"]["chatgpt"]
print(row["model"], row["target_found"], row["target_position"], row["total_citations"])
for c in row["citations"]:
    print(c["position"], c["domain"], c["title"])

Try it in the playground first if you want to see the JSON. Our visibility scoring guide explains the 0 to 100 number.

What to know before you build

• Every key is present on every response. An engine that cited nothing returns citations: [] and total_citations: 0.
• Zero and null differ. visibility_score: 0 means cited by nobody. null means you sent no domain.
• Compare on url_normalized, not url.
• Store the model label beside each result, so a label change explains a jump in your chart.
• Refunds. The landing page says a run in which every engine failed is refunded and a completed run is charged.

Limits and unknowns

• We did not test anything. Every competitor figure comes from its published page on 2026-10-23. Prices change, so treat them as dated.
• Pages we could not read. Peec’s plan prices, and Perplexity’s Sonar answer rates and free credits.
• Tools we left out. Hall’s site now says it is part of Tracksuit with a waitlist, so we could not confirm its free plan. We did not read Goodie, Athena or others, and omitting them is not a verdict.
• Units differ. A “prompt” in one dashboard is not a “check” in another, and we could not confirm how often each runs.
• Answers vary. AI answers change from run to run and between accounts and locations, so judge trends over repeated runs, not one result.
• Minimum top-ups. New Serpent accounts start at a $10 deposit. Once lifetime deposits reach $50, the minimum for further top-ups is $100, per the pricing page.
• Our interest. Serpent sells one of these options and we wrote the post. Every sum is shown, so you can redo any of it from the linked pages.

FAQ

Is there a free AI rank tracker?

Partly. HubSpot, Ahrefs, Trakkr and FixAEO offer free one-off checks, and several dashboards give 7 to 14-day trials. We found no free tool that tracks prompts daily with history across several engines. Ongoing tracking needs a paid plan or an API.

Is there a free ChatGPT rank tracker?

Only as a snapshot. Trakkr’s free check needs no login and runs on an OpenAI model with no web search, and HubSpot’s grader reports ChatGPT, Perplexity and Gemini from training data. Neither stores history.

What is the cheapest way to track AI visibility with an API?

Among the options we read, Serpent’s Growth and Scale rates are low: $2 and $1 per 1,000 single-engine checks, or $4 and $2 for four engines. DataForSEO lists $1.20 per 1,000 pages on its standard queue, for ChatGPT and Gemini only. Dashboards cost more per check.

Why do free AI visibility checkers show different results?

They measure different things. One free check runs a single model with no web search, another reports from training data, and answers also vary between runs, accounts and locations. Our earlier repeatability study (August, not a new test) saw citation counts change in 15 of 40 engine pairs. Compare only runs that used the same method.

Do I need a dashboard tool or an API?

Choose a dashboard if you want charts, alerts and no code. Choose an API if you want the answer text, cited sources and position in JSON for your own reports or agents. An API costs less per check, but you build the storage and charts yourself.

Sources read on 2026-10-23

• HubSpot: AI Search Grader
• Ahrefs: Brand Radar
• Trakkr: free AI rank tracker
• FixAEO: AI rank tracker
• Profound: pricing
• Otterly.AI: pricing
• Semrush: AI Visibility pricing
• Rankscale: pricing
• Writesonic: pricing
• Scrunch: pricing
• Peec AI: pricing
• Peec AI help: API cost
• DataForSEO: LLM Scraper pricing
• SE Ranking: API

Peec’s pricing page printed no plan amounts, so those figures are not verified. Related reading on Serpent: the AI citation repeatability study.