Photopea AI Background Remover · Background removal
Benchmark browser-native image suite powered by WebAssembly, delivering free signup-free 2048×2048 watermark-free transparent PNGs; watching a 30s rewarded ad unlocks exceptional curly hair matting precision (99.7% recall).
US-West (California, AS2914)
No native export attached; page screenshot evidence only
Subsequent pricing and service checks are shown separately without altering historical scores.
One run does not establish all complex scenes, untested paid maxima or long-term uptime.
Plans & historical pricing
Photopea Free and Premium pass structure · 2026-10-30
PSD/AI/XD/PDF editing, AI background removal, adjustment layers, bitmap vectorizer
No ads, one-time payment with no auto-renewal, reverts to free on expiry
Full-year ad-free experience, one-time payment without recurring billing
3 plan / billing-period records, all listed; — means that period is not recorded.
Free tier grants 100% feature access with ads; Premium passes are one-time non-recurring purchases (€8/30d, €15/90d, €50/365d) returning to free upon expiry.
Cancellation, sources & verification record
Cancellation and refund terms unverified
Photopea Official Premium Pricing Modal ↗ · 2026-10-30 Capture 1 ↗
Pricing recorded on the test date · 2026-11-03
Free with ads / Ad-free passes from €8/30 days (non-recurring)
Freemium (Free full-res export + non-renewing day passes)
Historical context only; not current purchase guidance.
Delivery & access
- Free delivery record
- 2048 × 2048 (4.96 MB original RGBA PNG)
- Watermark
- Completely watermark-free
- Free allowance
- Free users watch a single rewarded video ad to complete full-resolution AI export
- Account requirements
- Completely usable without login
- Maximum specification record
- 2048 × 2048 (4.96 MB original RGBA PNG)
May include advertised capabilities; this does not establish a tested paid tier.
- Batch & API
- Full keyboard shortcuts, PSD/AI import/export, action scripts, and blend modes
- Manual refinement
- Not recorded or not verified
Tested result & sample
Photopea exports full 2048×2048 transparent PNG client-side with outstanding hair preservation (fg_recall = 99.74%, bg_leak = 0.026%), generating an editable layer mask.
- Edge error MAD
- 0.112 · Lower is better
- Global error SAD(k)
- 44.707 · Lower is better
- Foreground retention
- 99.740% · Higher is better
- Background residue
- 0.026% · Lower is better
These historical measurements describe alpha error at edges (MAD), total alpha error / 1000 (SAD), foreground retention and background residue. Low-resolution outputs are aligned to the reference under the original protocol; this does not imply a full-resolution export.
- Fixture
- anchor_bgremove_v1
- Recorded environment
- US-West (California, AS2914)
- Local processing record
- 15 s
- End-to-end reference
- 50 s
Network wait includes round trips, queuing and polling error, not pure inference time. Local timings depend on the device; repetition counts are not standardized in legacy records.
No native export is attached. Screenshots show the visible page, but do not independently establish export properties or file validity.
Experience, rules & trust
- Disclosure
- Explicitly informs user of ad requirement upon clicking Remove BG
- Interaction
- Not recorded or not verified
- Entity record
- Ivan Kutskir (Photopea)
- Origin
- Czech Republic
- Domain history record
- 2013–present
- Third-party tracking record
- Google AdSense · Google Analytics
- External information
- Not recorded or not verified
- Processing
- Client-side WebAssembly / Canvas offline neural network inference
- Retention statement
- Data stays entirely within local browser memory with zero server uploads
A recorded statement does not verify deletion or all data-processing behavior.
Verifiable audit trail & evidence
Four SHA-256-verified stages link the benchmark fixture, same-run interface screenshots, native exported artifact and archived pricing sources.
anchor_bgremove_v1 · Inspect fixture & file inspector →
Captures entry rules, result view and export disclosure.
Page screenshot evidence only
Screenshots, native output and official claims support different conclusions. Evidence grades are not product safety certifications.
Evaluation basis & full scores · Overall 92.9/100
Overall = honesty & reputation × 30% + capability × 70%. Component maxima follow this report’s rubric. List scores are rounded for display; original precision is retained here.
Scoring version 2.3.1 · Test fixture anchor_bgremove_v1
Honesty & reputation · 96/100 · 30%
| Criterion | Score | Rationale | Basis |
|---|---|---|---|
| Pricing transparency | 34 / 35 | Explanation unavailable | Evidence A → |
| Interaction restraint | 27 / 30 | Explanation unavailable | Evidence A → |
| Background & reputation | 20 / 20 | Explanation unavailable | Evidence A → |
| Engineering & privacy | 15 / 15 | Explanation unavailable | Evidence A → |
Capability · 91.5/100 · 70%
| Criterion | Score | Rationale | Basis |
|---|---|---|---|
| Task quality | 58.5 / 65 | Transition band error mad_band=0.112 · Global SAD(k)=44.707 · Foreground recall 99.74% / Background leak 0.026% | Evidence B → |
| Deliverable specifications | 20 / 20 | True Alpha channel (8/8); Max deliverable spec: 2048 × 2048 (4.96 MB original RGBA PNG) | Evidence B → |
| Processing speed | 13 / 15 | Network-observed server wait: 15s (API mode: client_wasm_neural_matting; network upload 0s / end-to-end 50s shown for reference only, never penalized) | Evidence B → |
Legacy scores may include advertised maximum capabilities. Read tested output separately from paid capabilities. Records without matching tiers and scope are not ranked together.
Revisions & corrections
- 2026-11-03
Completed real-machine benchmark evaluation and 3-node screenshot verification under RealTest Protocol v2.3.