Real cost of AI watermark removal — browser vs SaaS vs self-hosted GPU
If you only have time to read three lines: a Telea-based browser tool costs roughly $0.0000003 per image (the electricity to power your phone for a fifth of a second). A typical SaaS watermark tool at $9.99/mo costs $0.005 per image if you use it 2,000 times a month, or $0.05 if you only use it 200 times. Self-hosting LaMa on an A100 at $1.50/hr costs $0.00008 per image in GPU time alone, plus bandwidth, plus the 50 MB weight download that every user pays for once.
Three approaches, three cost stacks
When you remove a watermark you pay for something. What you pay for and who you pay depends on where the work happens:
- Browser tool: the work runs in your browser on your phone or laptop. You pay for the electricity to power your device, and nothing else.
- SaaS subscription: you upload your image to a vendor’s server. The vendor runs the model. You pay a monthly fee that covers their compute, their bandwidth, their storage, their support inbox, and their profit.
- Self-hosted GPU: you rent a virtual GPU from a cloud provider, run the model yourself, and charge yourself the GPU-hour rate. You pay for the GPU time, the egress bandwidth, and the disk to hold the model weights.
Same image, same model underneath (sometimes), three completely different price tags. We break each one down below.
1. The browser tool: electricity only
The site you are reading runs Telea 2004 in your browser. The full pipeline (decode + inpaint + encode) for an 800x600 image on a 2024 mid-range phone takes about 250 ms. Phone SoC power draw under sustained compute is roughly 4-6 W; we will use 5 W as the midpoint.
Per image:
- Energy = 5 W × 0.250 s = 1.25 J = 0.000347 Wh
- At the US average residential electricity rate of $0.16 / kWh (US EIA, mid-2026): $0.0000000556 per image.
- Rounded: $0.0000001 per image, or about one hundred-thousandth of a cent.
At any human scale of usage — a thousand images, a hundred thousand images, a million images — the electricity is rounding noise in your monthly phone bill. The actual cost to you is zero; the actual cost to the grid is one ten-thousandth of a cent. The operator of a browser tool pays nothing per image: bandwidth is the page itself, compute is the user’s device.
Source: US Energy Information Administration Average Retail Price of Electricity, residential sector, July 2026 release.
2. The SaaS subscription: what you actually pay for
The market has settled on roughly three price tiers for watermark-removal SaaS. Names vary; numbers do not. We use ranges we have seen across the segment in 2026:
| Tier | Monthly price | Typical image cap | Per image if you max it | Per image if you use 200/mo |
|---|---|---|---|---|
| Starter | $4.99 | 500 images | $0.010 | $0.025 |
| Pro | $9.99 | 2,000 images | $0.005 | $0.050 |
| Business | $29.99 | 10,000 images | $0.003 | $0.150 |
| Pay-per-image | — | — | $0.10 - $0.50 per image | — |
That per-image column is what matters and most users miss it. A $9.99/mo subscription is not “unlimited watermark removal”; it is “$9.99 for up to 2,000 images, and $0.005 per image if you use it fully, and $0.05 per image if you only use 200 of them”. The break-even for the Business tier is roughly 600 images per month. Below that, you are overpaying.
What the SaaS price covers that the browser tool does not:
- Server compute: 200–500 ms of an A100-class GPU per image, which is roughly $0.0001 of raw GPU time at $1.50/hr A100 spot pricing. The vendor pays this whether you max the tier or not.
- Bandwidth: 0.5-2 MB of image data up, 0.5-2 MB of cleaned image down. At $0.05/GB egress (typical US-east cloud rate) that is $0.0000001 per image. Effectively free.
- Storage: 30 days of retention is typical. A few TB for a small SaaS, costing $20-100/mo on object storage.
- Support, refunds, payment processing: the biggest line item. A SaaS with 1,000 subscribers spends more on support than on compute.
- Profit: a sustainable SaaS in this segment runs at 50-70% gross margin. Half of your $9.99 is gross profit, not cost.
So the per-image cost of a SaaS is dominated by shared costs (support, profit) spread across active users, not by the per-image GPU time. If you max your tier, you are getting closer to a fair price. If you do not, you are subsidising the heavy users.
3. The self-hosted GPU: paying for the machine, not the image
You rent a virtual GPU. You run the model yourself. You pay the cloud. This is the model a serious image-processing shop picks when they are handling 100k+ images per month or have compliance reasons to keep data in-house.
Per-image compute cost on common GPU options in 2026 (spot pricing, us-east-1 equivalents):
| GPU | Hourly rate (spot) | Time per image (LaMa) | Per-image GPU |
|---|---|---|---|
| NVIDIA A100 40GB | $1.50/hr | 200 ms | $0.0000833 |
| NVIDIA A10G 24GB | $0.55/hr | 500 ms | $0.0000764 |
| NVIDIA L4 24GB | $0.40/hr | 700 ms | $0.0000778 |
| NVIDIA T4 16GB | $0.20/hr | 1500 ms | $0.0000833 |
The per-image GPU cost converges around $0.00008 regardless of which card you pick, because the slower cards are cheaper per hour. The interesting number is not per-image but per-month.
Realistic monthly bill for a self-hosted LaMa shop processing 100,000 images:
- GPU: A10G 24/7 at $0.55/hr × 720 hr = $396/mo
- Compute instance (CPU + RAM + storage): ~$80/mo
- Egress: 100,000 images × 1 MB × $0.05/GB = $5/mo
- Model weight storage (50 MB, negligible): ~$0.01/mo
- Load balancer, queue, monitoring: ~$40/mo
- Operator time: 2 hours/mo at $50/hr for a competent engineer = $100/mo
- Total: ~$621/mo, or $0.0062 per image.
If your usage is bursty (10,000 images on Monday, nothing Tuesday through Sunday), you can use a serverless GPU product like Replicate or Modal, which bills per second. At Replicate’s published LaMa price of $0.0007 per image (serverless cold-start included), 100,000 images is $70/mo of pure compute, plus whatever orchestration you build around it.
Putting the three together
The break-even is the question that matters. The browser tool is free; SaaS and self-host both have fixed costs. We compare per-image cost across realistic volumes:
| Volume (images/mo) | Browser tool | SaaS Pro $9.99 | Self-hosted A10G 24/7 | Replicate LaMa |
|---|---|---|---|---|
| 100 | $0.00000001 | $0.10 | $6.21 (over 100x) | $0.07 |
| 1,000 | $0.0000001 | $0.010 | $0.62 (over 100x) | $0.0007 |
| 10,000 | $0.000001 | $0.001 | $0.062 | $0.0007 |
| 100,000 | $0.00001 | $0.0001 | $0.0062 | $0.0007 |
| 1,000,000 | $0.0001 | $0.00001 (at cap) | $0.000621 | $0.0007 |
The browser column is mostly a joke at these volumes — it is literally less than the rounding error of any other line. SaaS at the Pro tier crosses below $0.001 per image only at 10k images per month or more. Self-hosted wins at scale because the GPU is paid for whether you use it or not, but you have to actually use it 24/7 to get the per-image number above.
The interesting number is the crossover between SaaS and self-hosted. SaaS at $9.99/mo for 2,000 images is $0.005 per image. Self-hosted at $621/mo for 100,000 images is $0.0062 per image. The crossover volume is roughly 50,000 images per month. Below that, SaaS. Above that, self-hosted.
The hidden costs nobody puts on the page
Every option has costs that the price page does not show:
- Browser tool hidden cost: your time and your device. A browser tool runs on your phone, so it costs you nothing in cash but it does cost battery, it does cost a few minutes, and it does keep a record on your device. The privacy upside is real; the convenience downside is real too.
- SaaS hidden cost: data leaving your device. You are uploading your image to a third party. That image might be a draft of a paid stock photo you licensed. It might be a scanned ID. It might be a screenshot of your bank statement. The SaaS contract usually grants them a licence to use the image for “service improvement”. The browser tool does not have this cost.
- Self-hosted hidden cost: you are now a sysadmin. GPU drivers, CUDA versions, model weight updates, queue depth, autoscaling, model serving (Triton / vLLM / Ray Serve), monitoring, alerts. None of this is in the $621/mo number. If your engineer costs $100/hr and spends 4 hours a month on this, the real per-image cost doubles.
- Self-hosted hidden cost: cold start. A GPU that has been idle for 10 minutes pays a 10-15 second cold start to load the model. Bursty workloads pay this penalty per burst. Spot instances can be reclaimed mid-job. Self-hosted is not a fire-and-forget option.
- SaaS hidden cost: feature lock-in. If your SaaS goes down, you cannot process images. If your SaaS gets acquired and shut down, you lose the model. If your SaaS changes its terms, you have to migrate. Open formats and open code survive vendor changes; subscriptions do not.
What the right answer looks like
There is no single right answer. Three honest recommendations:
- Less than 1,000 images per month, on simple backgrounds: use a browser tool. The cost is rounding noise, the privacy is intact, and the quality is fine on flat/gradient images. This site is one such tool; others exist.
- 1,000-50,000 images per month, mixed backgrounds: SaaS Pro tier is the simplest answer that costs less than a part-time hire. Budget $50-300/mo depending on volume. Pick a vendor that lets you export your images and the cleaned version for at least 30 days, so you can move if pricing changes.
- 50,000+ images per month, compliance-sensitive, or custom model needs: self-host on a single A10G or L4, run it 24/7, accept the sysadmin cost. Use a managed Kubernetes GPU product if you do not want to roll your own queue.
FAQ
Why is the browser tool not $0 exactly?
It is $0 to the operator of the tool and effectively $0 to the user. The electricity cost of running the inpainter on a phone is less than one ten-thousandth of a cent per image, which is below the resolution of any monthly bill. We use the rounded figure $0.0000001 to make the comparison honest, not to suggest you will ever see this on a bill.
Do I pay for bandwidth in the browser tool?
Yes, but it is fixed, not per-image. The browser tool’s HTML, CSS and JS are a few hundred KB. They load once when you visit the page. After that, every image you process uses zero additional bandwidth, because the inpainting runs on your device. A SaaS tool charges per-image because every image is an upload to their server.
What about free tiers?
Free tiers usually mean: 3-5 images per month, watermarked output, or rate-limited to 1 image per minute. They exist to convert you to a paid tier, not to be a sustainable way to process images. A free tier that is actually free forever (like this site) is paid for by ads, by donations, or by the operator’s day job. Read the small print to find out which.
Does the GPU choice really not matter per image?
It matters less than people think. A faster GPU costs more per hour but finishes each image faster; a slower GPU costs less per hour but takes longer. For LaMa at the typical 200-1500 ms range, the per-image cost is dominated by the “GPU idle cost amortised over actual work” equation, and that converges to $0.00008-ish for any reasonable card. The choice between A100 and T4 is mostly about which workloads you want to handle, not about per-image cost.
What is the cheapest legal way to remove a watermark?
If you have the right to edit it, a browser tool is the cheapest, with the privacy bonus. If you do not have the right to edit it, the cheapest answer is still no, because infringement damages are not in the cost model above.
Methodology and sources
Electricity: US EIA Average Retail Price of Electricity, residential, July 2026 release. SaaS pricing: observed across the segment in 2026, anonymised. GPU spot pricing: AWS, GCP, Lambda Labs and RunPod published rates for us-east-1 equivalents in Q3 2026. Runtimes: 200 ms for LaMa on an A100 from the official LaMa repo benchmark, 500-1500 ms on A10G/L4/T4 measured on the Replicate hosted LaMa model. Operator time estimate for self-hosted is a midpoint between 1 hr/mo (well-run, autoscaled, no surprises) and 8 hr/mo (managed badly with weekly incidents).