Site icon Filestack Blog

Is the Filestack Free Plan Enough for Your Project

The Filestack free plan is enough for a prototype, portfolio piece, internal tool, or side project moving a few hundred files. It stops being the right fit when the project needs AI work on files, more team members, or production-scale volume.

The quotas below are from the pricing page as of 31 July 2026. The capability examples combine the current plan table with requests run against a free key.

Key takeaways

What the free plan actually runs

The free plan includes dozens of core processing operations. Our July 2026 test key ran the 62 operations grouped below:

Category Count What is in it
Image transformation 25 resize, crop, rotate, flip, blur, pixelate, sharpen, watermark, collage, animate, compress, quality, modulate, ascii, and more
Picker sources 12 Google Drive, OneDrive, Box, Google Photos, Instagram, Facebook, webcam, video, audio, URL, image search
Decorative 7 border, circle, polaroid, rounded_corners, shadow, torn_edges, vignette
Face detection 4 detect_faces, crop_faces, blur_faces, pixelate_faces
Format conversion 4 jpg, webp, pdf, txt
Document 3 pdfinfo, pdfconvert, metadata
Delivery 2 cache, fallback
Code minification 2 minify_js, minify_css
Archive, storage, capture 3 zip, store, urlscreenshot

The free plan covers the core image editing api tasks, PDF text extraction, and facial detection. That includes the blurring and pixelating variants. Parameters for every task are in the processing API reference.

Check it against your own project

That table is our answer for everyone. Run this to get the one for you. Upload a file, then fire off the tasks your project actually calls:

HANDLE="YOUR_IMAGE_HANDLE"

for task in resize=width:400 crop=dim:%5B0,0,400,400%5D output=format:webp compress \
            crop_faces=faces:1,mode:thumb watermark=file:$HANDLE,size:30; do
  printf '%-40s ' "$task"
  curl -s -o /dev/null -w '%{http_code} %{size_download}B\n' \
    "https://cdn.filestackcontent.com/$task/$HANDLE"
done

No API key goes in those URLs, because the handle already identifies the application. A 200 with a size different from your original means the task ran. Every task and its parameters are in the processing API reference.

It takes a minute, and the full script and how to read it goes through every line of the output.

What the higher plans add

The AI-backed operations, and they are worth knowing by name because each one maps onto a whole category of product: optical character recognition, image tagging, safe-for-work classification, captioning, copyright checking, document detection, image and text sentiment, smart cropping, enhancement, upscaling, red-eye removal, and video conversion.

We split the tiers there because those operations run trained models rather than arithmetic, which makes the boundary easy to predict. If an operation has to understand what is inside a file, it is on a higher plan. If it changes the file’s shape, size or format, it is on the free one.

The distinction worth getting right early is inside document work. Extracting text from a PDF that already has a text layer is a free operation and takes about a quarter of a second. Reading text off a scan or a photograph is optical character recognition, which is on the higher plans. Those two look identical from the outside, so work out which one your documents need before you plan around them.

The quotas

Free Start, $69 a month
Price $0, free forever $69 billed monthly
Uploads 500 20,000
Bandwidth 1 GB 75 GB
Transformations 1,000 50,000
Storage 1 GB 50 GB
Team members 1 up to 5

Grow and Scale sit above that at $199 and $379 a month, and Enterprise is negotiated.

Transformations are the line people misjudge, so let me be precise about how we count them. Every distinct transformation URL is one, which means a product page rendering four image sizes uses four. What pulls it back is caching. A resized file comes back with a max-age of about 30 days and the CDN serves the repeat requests from an edge node without touching the origin, so 1,000 transformations is 1,000 distinct URLs rather than 1,000 page views. That is why we point free-plan users at the caching behaviour in the file delivery walkthrough early. Choosing when to convert to webp is the other lever, and it works on bandwidth rather than on transformations.

The signs you have outgrown it

The numbers matter less than what each counter measures, since three of the four reset monthly and one does not.
Any one of these means the Start plan is the right next step:

Most projects meet one of these on their way up rather than all at once, and the first one you meet is usually the reason to move.

What to do when one of them becomes true

Do the cheap things first, because they are work you want on any plan. Let caching do its job instead of bypassing it the way you did in development, resize before converting format, and serve WebP where the browser accepts it. Those three push the bandwidth and transformation lines a long way, and the wider approach is in the file delivery workflow guide.

Then move up for the reason that actually applies. If it is an AI operation, you already know which one and the plan follows from that. If it is quota, Start multiplies your uploads forty times and your bandwidth seventy-five times for $69 a month.

The short version is that we made the free plan generous on capability and deliberately narrow on volume, seats and AI. Start there, build the thing, and move up at whichever of those three you reach first. Create a free account and run your own operations against it, because the answer for your project is about thirty minutes of testing away.

Exit mobile version