ApiFlash publishes four plans and the capability list on all four is identical. Free is 100 screenshots. Lite is 1,000 for $7, which is the cheapest paid rung anyone in this category sells. Medium is 10,000 for $35. Then the ladder jumps straight to Large: 100,000 for $180. There is nothing between the last two, so the ten thousand and first screenshot of the month costs $180.
Our ladder publishes nine rungs between 3,000 and 200,000, so 15,000 screenshots is $49.99 rather than $180, and 30,000 is $59.99. Below 10,000 they are cheaper and section 03 concedes those rows without hedging. The other difference is what arrives alongside the capture: a schedule, change detection, alerts, and a 48-month archive, none of which appear anywhere on their site.
ApiFlash pricing verified 28 July 2026 against the pricing section of apiflash.com.
per month
Medium stops at 10,000
so this is the next rung
per month
a rung that fits
plus the archive
Their Medium plan covers 10,000. The next one covers 100,000. Anything between the two buys the larger plan, which is what a four-rung ladder does to the middle of a market.
Their pricing is admirably plain, so it is worth stating exactly. Free: 100 screenshots. Lite: 1,000 for $7 a month. Medium: 10,000 for $35. Large: 100,000 for $180. Custom Enterprise above that, with no published number attached to it.
All four cards publish the same seven capabilities: PNG, JPEG and WebP output, full-page screenshots, mobile screenshots, ad blocking, cookie-banner hiding, and export to S3. The only thing that changes as you climb is the screenshot count and the adjective in front of the word support. That is an honest model and we would rather credit it than pretend otherwise.
The problem is the shape of the ladder rather than its honesty. From Medium to Large the volume multiplies by ten while the price multiplies by five, and nothing sits between them. Use 12,000 screenshots a month and you buy 100,000. Use 40,000 and you buy 100,000. The entire middle of this market pays the top of their ladder.
Ours publishes nine rungs from 3,000 to 200,000, which is the whole reason 15,000 screenshots is $49.99 here and $180 there. Below 10,000 the comparison runs the other way, and section 03 says so in two rows rather than leaving them out.
| ApiFlash plan | Shots | Per 1,000 | Support |
|---|---|---|---|
| Free $0 | 100 | — | none |
| Lite $7 / mo | 1,000 | $7.00 | Basic |
| Medium $35 / mo | 10,000 | $3.50 | Priority |
| Large $180 / mo | 100,000 | $1.80 | Premium |
| Custom Enterprise no published price | Custom | — | Top |
Two rows go to them: the $7 entry plan and the price of exactly 100,000 screenshots. Their model is also simpler than ours, which counts for something a table cannot score.
| Capability | PagePixels | ApiFlash |
|---|---|---|
| What the product is | Scheduled capture, change detection, and an archive | A capture endpoint you call |
| Published plan rungs | Nine, from 3,000 to 200,000 screenshots | Four: 100, 1,000, 10,000, 100,000 |
| Entry paid price | $16.99 / mo for 3,000 screenshots | $7 / mo for 1,000. This is theirs |
| Free plan | 25 screenshots / mo, every feature, 48-month archive | 100 screenshots / mo, every feature, no archive |
| Cost of 15,000 screenshots | $49.99 / mo | $180 / mo. Medium stops at 10,000 |
| Cost of 100,000 screenshots | $199.99 / mo | $180 / mo. This is theirs |
| Cost of 200,000 screenshots | $399.99 / mo | Custom Enterprise. No published price |
| What changes between tiers | The screenshot count only | The screenshot count and the support level only |
| Recurring schedule | Built in, 5 minutes to monthly, every plan | Not listed on their site. Bring your own cron* |
| Change detection | Included on every plan, with region exclusions | Not listed on their site* |
| Alerts when something moves | Email, Slack, and webhooks on every plan | Not listed on their site* |
| Where captures are stored | Our archive, dated, 48 months, every plan | Your own S3 bucket, on every plan including free |
| Cost of retention | $0. Included, with no cap on archive size | Whatever your bucket costs, for as long as you keep it* |
| Webhooks | Included on every plan including free | Not listed on their site* |
| Slack and no-code delivery | Slack, Zapier, Make, n8n, Power Automate, Dropbox, all plans | Not listed on their site* |
| Export direct to your S3 | Not offered. Dropbox, Slack, and webhooks instead | Every plan including free. This is theirs |
| Choice of capture location | 150+ countries, states, and cities, every plan | IP geolocation listed under Custom Enterprise* |
| Managed proxies | Not sold separately. Locations are included | Custom Enterprise only. No published price |
| Full page, mobile, ad and banner blocking | Every plan | Every plan including free. A genuinely clean model |
| Pages behind a login | Every plan. 13 action types: fill, select, click, submit, wait | No ordered action sequence listed on their site* |
| MCP server for agents | Hosted, OAuth 2.1, the whole API as tools | None found |
| Top published tier | $399.99 / mo for 200,000 screenshots | $180 / mo for 100,000. Above that is Custom |
ApiFlash pricing and feature lists verified 28 July 2026 against the pricing section of apiflash.com. All four plan cards publish the same seven capabilities, so what changes across their ladder is volume and support level rather than features. This page is monitored for changes by PagePixels, so when their pricing moves, this table gets corrected.
Monthly billing on both sides, and the only variable is how many screenshots you take. Under 10,000 they are cheaper. Between 10,000 and 100,000 the gap in their ladder does all the work. At exactly 100,000 they are cheaper again.
| Workload | Shots / mo | ApiFlash | PagePixels | Difference |
|---|---|---|---|---|
| A thumbnail per signup | 1,000 | $7 (Lite) | $16.99 | They win by $9.99. Our smallest paid plan is 3,000 |
| Preview images for a small SaaS | 3,000 | $35 (Medium) | $16.99 | We win by $18.01. Lite stops at 1,000 |
| 330 pages, captured daily | 10,000 | $35 (Medium) | $29.99 | We win by $5.01. This is their best-value rung |
| 500 pages, captured daily | 15,000 | $180 (Large) | $49.99 | We win by $130.01. This is the gap in one row |
| 42 pages, captured hourly | 30,000 | $180 (Large) | $59.99 | We win by $120.01 |
| 97 pages, captured hourly | 70,000 | $180 (Large) | $139.99 | We win by $40.01 |
| 139 pages, captured hourly | 100,000 | $180 (Large) | $199.99 | They win by $19.99, at full use of their top plan |
| 69 pages, every 15 minutes | 200,000 | Custom | $399.99 | Nothing published above 100,000 on their side |
Plan price divided by the screenshots you actually take, at 15,000 a month.
| Plan | Cost per thousand screenshots |
|---|---|
| ApiFlash Large, buying 15,000 | $12.00 |
| ApiFlash Lite rate, for reference | $7.00 |
| ApiFlash Medium rate | $3.50 |
| ApiFlash Large at a full 100,000 | $1.80 |
| PagePixels 15,000, at $49.99 | $3.33 |
Their rate at full utilisation is excellent. Their rate at 15,000 is worse than their own entry plan, because you are buying 100,000 in order to use 15,000. Ours is $3.33 because there is a rung there.
Two rows, and the first is not close. A thousand screenshots a month is $7 there and $16.99 here, because their smallest paid plan is 1,000 and ours is 3,000. If a thousand is genuinely your ceiling, buy from them.
The second is exactly 100,000, where Large is $180 and our 100,000 plan is $199.99. A difference of $19.99 a month. If your volume sits on that number and you already run a scheduler and a storage bucket, they are the cheaper way to render.
There is a third thing no table scores. Their product is one endpoint with seven options and the same list on every tier. Ours carries a scheduler, change detection, an archive, and integrations, and if you do not need those, you are paying for them.
ApiFlash is a capture endpoint and it does not pretend to be anything else. You send a URL, it sends back an image, quickly and reliably, off current Chrome on AWS Lambda. Nothing on their plan cards, in their feature section, or on their homepage fires a capture on an interval, compares one capture to the previous one, or tells you what moved.
That is a legitimate product and for plenty of jobs it is the right one. Thumbnails and social preview images do not need a diff. The moment the job becomes watch this page and tell me when it changes, the missing parts are the ones that take longest to build: a scheduler that survives a deploy, storage with a retention policy, a comparison that ignores the regions that always move, and an alert that reaches a person.
One row in the table below goes to them, and it is the last one. Export straight to your own S3 bucket is on every ApiFlash plan including the free one. We deliver to Dropbox, Slack, and any webhook, and we keep the archive ourselves, but a direct write into your bucket is a thing they do and we do not.
| Capability | ApiFlash | Ours |
|---|---|---|
| A recurring schedule | build it not on their site | free |
| Change detection between captures | build it not on their site | free |
| An alert when something moves | build it not on their site | free |
| Storage with a retention policy | your S3 bucket is yours | free |
| Webhooks on capture | build it not on their site | free |
| Login and multi-step actions | build it not on their site | free |
| Choice of capture location | Custom Enterprise | free |
| Export direct to your own S3 | $0 every plan | not offered |
This section is usually about API access, and here it is not, because ApiFlash is an API and a good one. One documented endpoint, code examples in seven languages, and a fast Chrome-on-Lambda backend that eleven thousand businesses point at. We are not going to argue with them about rendering.
What we sell is the layer above it, and ours is on the free plan in full: create and update configs on a schedule, trigger a capture on demand, read change notifications, list the archive. So is the hosted MCP server, which means an agent can set up a watch today and read the diff next month without you writing any glue.
If you are already running that layer on your own cron and your own bucket, $180 for 100,000 renders is a fair price and you should take it. If you are about to write it, the cost of writing it is the thing this page is asking you to put a number on.
ApiFlash gates nothing by feature, and that deserves saying plainly: their free plan carries the same capability list as their $180 one. What differs is that the list is seven items long. Here the list below is the free plan, and the paid plans differ from it only in how many screenshots you get.
Published plans at 3,000, 10,000, 15,000, 30,000, 50,000, 70,000, 100,000, and 200,000 screenshots a month. Their ladder goes 1,000, 10,000, 100,000, so everything in the middle of the market buys the top rung.
Set an interval from five minutes to monthly and the capture fires without you. Nothing they publish schedules anything, so on that side the interval is your cron job, your queue, and your uptime.
Every capture stored, dated, and openable for four years, with no cap on archive size and no storage line on the bill. Their captures land in your S3 bucket, where the retention policy and the invoice are yours.
Compare this capture to the last one, exclude the regions that always move, and get told when something real changes, by email, Slack, or webhook. That comparison is code you write and maintain on their side.
Choose the country, state, or city a page is captured from, on any plan. IP geolocation and managed proxies appear on their Custom Enterprise card, which carries no published price.
Thirteen action types run in the order you list them: log in, fill the form, choose the dropdown, click through, wait for the element, then capture. Nothing on their site lists an equivalent ordered sequence.
Four cases, and the first two are pure arithmetic. If a thousand screenshots a month is your ceiling, or if your volume sits exactly on a hundred thousand, they are cheaper and you should buy from them.
Their Lite plan is $7 for 1,000 and it is the cheapest paid rung anyone in this category sells. Our smallest paid plan is $16.99 for 3,000, so at that volume they are less than half the price and we are selling you headroom you will not use. Buy from them.
Large is $180 for 100,000 screenshots. Our 100,000 plan is $199.99. That is $19.99 a month in their favour at full utilisation of their top published tier, and $1.80 per thousand is a good rate by any measure in 2026.
If you run your own cron, your own S3 lifecycle rules, and your own diffing code, then most of what we charge for is infrastructure you already operate. Paying us for it twice is not a good trade. Their render endpoint plus your existing stack is the cheaper answer.
Four plans, seven capabilities, the same list on the free tier as on the $180 one, and nothing gated except how fast someone answers your email. If all you need is a thumbnail, that simplicity has real value and our feature list is mostly noise to you.
Count your monthly volume before anything else. Under 10,000 screenshots this migration probably costs you money, and section 03 has the two rows where it does. Above 10,000 the arithmetic turns, and what you are really moving is not the render call but the scheduler and the bucket around it.
Honest note: their export to S3 is a real feature and we do not match it. If a direct write into your own bucket is load-bearing for you, that is a reason to stay. What builds here is kept in our archive for 48 months while the account stays active, at no extra storage cost.
It depends entirely on volume, and the answer flips twice. Up to 10,000 screenshots a month they are cheaper: $7 for 1,000 and $35 for 10,000 against our $16.99 and $29.99. Between 10,000 and roughly 90,000 we are cheaper, often by a lot, because their next rung after Medium is $180 while ours climbs in six steps and 15,000 is $49.99. At exactly 100,000 they are cheaper again, $180 against $199.99. Find your number before you pick a side.
Because their Medium plan covers 10,000 and the next plan covers 100,000, with nothing between them. Any volume in that range buys the larger plan. At 15,000 that works out to $12.00 per thousand, which is more than their own $7 entry plan charges per thousand.
Nothing on their plan cards, feature section, or homepage schedules a capture. It is a capture endpoint: you call it, it returns an image. Firing it on an interval means your own cron, a Zapier plan, or a worker you keep alive. Scheduling is built into every PagePixels plan including the free one, from five minutes to monthly.
Not that their site lists. It returns the image. Comparing it to yesterday's, deciding what counts as a real change rather than a rotating banner, and getting that to a human is code you write and then maintain. It is included on every plan here, with region exclusions so the parts that always move do not page you.
In your own S3 bucket, through their export feature, which is on every plan including the free one. That is a genuinely good feature and we do not match it. It also means the bucket, the retention policy, and the storage bill are yours to run. Our archive is included: every capture dated and kept for 48 months while the account stays active, with no cap on size.
Their plan cards list IP geolocation and managed proxies only under Custom Enterprise, which publishes no price. Their homepage also shows an example rendered with an ip_location parameter, so the capability exists and it is the tier boundary that is unclear. Ask them. Ours is 150+ countries, states, and cities on every plan including free.
Three things, all real. Their $7 plan for 1,000 screenshots is the cheapest paid rung in the category. At exactly 100,000 screenshots they are $19.99 a month cheaper than we are. And export straight to your own S3 bucket is on every one of their plans, including free, which we do not offer at all.
No, and it is worth crediting. Their free plan lists the same seven capabilities as their $180 plan: PNG, JPEG and WebP output, full-page and mobile capture, ad blocking, cookie-banner hiding, and S3 export. What changes across their ladder is volume and support level, not features. Ours works the same way, at 200 screenshots a month.
Ours is hosted at mcp.pagepixels.com with OAuth 2.1 and exposes the whole API as tools, including the archive. We searched their site, their documentation, and general search on 28 July 2026 and found no MCP server. That is a negative search result rather than a vendor denial, so we re-check it quarterly.