Magic Eraser (Magic Studio) · Object removal
Login-free, ad-free brush removal and comparison are simple; free 600px versus Pro HD is shown only at download, limiting free detail.
US-West (California, AS2914)
600 × 600 · PNG · Open native file ↗
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
Current Chinese web CNY Free and PRO 3/6/12-month plans · 2026-10-20
Ads and Magic Studio logo
4 plan / billing-period records, all listed; — means that period is not recorded.
Pricing now loads: 3/6/12-month totals are CNY 149.99/249.99/239.99. The current 12-month promotion is preserved without multiplying rounded monthly prices. Free use has no count limit but is capped at 600px with branding. Paid pixel limits, renewal amounts and other regional quotes are not stated in this component.
Verification has gaps; unknown prices and quotas are not inferred.
Cancellation, sources & verification record
FAQ states no refunds except where legally required; a separate renewal-cancellation flow was not verified
Official plans and usage terms ↗ · 2026-10-20 Capture 1 ↗
Loaded CNY subscription pricing and FAQs ↗ · 2026-10-20 Capture 1 ↗ · Capture 2 ↗ · Capture 3 ↗ · Capture 4 ↗
Pricing recorded on the test date · 2026-10-17
Free 600px export; Pro ~$33/year (~$2.75/mo billed annually)
Free low-res tier + All-in-one Pro subscription + API
Historical context only; not current purchase guidance.
Delivery & access
- Free delivery record
- 600 × 600 (8.58% pixel area retained; 2048×2048 requires Pro)
- Attached file dimensions
- 600 × 600 · PNG
Read directly from the attached file; this does not establish free download availability at that size.
- Watermark
- Unobstructed main image at 600×600 free export
- Free allowance
- Unlimited free guest brush inpainting at 600×600 output resolution
- Account requirements
- No login required for 600×600 free tier; Email/Google login for Pro
- Maximum specification record
- 600 × 600 (8.58% pixel area retained; 2048×2048 requires Pro)
May include advertised capabilities; this does not establish a tested paid tier.
- Batch & API
- Batch editing and API supported on Pro plans
- Manual refinement
- Manual erase / adjustment recorded in the historical flow
Tested result & sample
Magic Erasers neon-green brush workflow cleanly erased all three occlusion signs in a single pass (~6.5s, 02_workspace.png, output.png), blending wood and foliage tones naturally; the main limitation is the 600×600 free resolution cap (unmasked_psnr_db = 24.89 dB).
- Fixture
- anchor_inpaint_v1
- Recorded environment
- US-West (California, AS2914)
- Network-observed server wait
- 6.5 s
- End-to-end reference
- 8 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.
Experience, rules & trust
- Disclosure
- Disclosed when clicking the top Download button
- Interaction
- No signup or login required for unlimited brush strokes, inpainting passes, and 600×600 free downloads.
- Entity record
- Magic Studio Inc.
- Origin
- USA / Global Remote Team (Founded 2021)
- Domain history record
- 2021–present
- Third-party tracking record
- Not recorded or not verified
- External information
- Not recorded or not verified
- Processing
- Client-side Canvas mask painting + Cloud GPU Inpainting
- Retention statement
- Temporary session images are purged periodically
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_inpaint_v1 · Inspect fixture & file inspector →
Captures entry rules, result view and export disclosure.
Official plans and usage terms (snapshot ↗) · Loaded CNY subscription pricing and FAQs (snapshot ↗)
Screenshots, native output and official claims support different conclusions. Evidence grades are not product safety certifications.
Evaluation basis & full scores · Overall 72.2/100
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_inpaint_v1
Honesty & reputation · 79.5/100 · 30%
| Criterion | Score | Rationale | Basis |
|---|---|---|---|
| Pricing transparency | 25.5 / 35 | Anonymous guests can upload the 2048×2048 test image directly into the brush canvas, paint over unwanted objects, and click Erase to inspect the result; clicking Download opens a split modal offering Pro high-res plans or a free "Download in Low Resolution" (600×600) button (02_workspace.png, 03_download.png). | Evidence A → |
| Interaction restraint | 24 / 30 | No signup or login required for unlimited brush strokes, inpainting passes, and 600×600 free downloads. | Evidence A → |
| Background & reputation | 18 / 20 | Magic Eraser (magiceraser.io / magicstudio.com/magiceraser) is one of the web’s original viral AI object removal tools, developed by Magic Studio Inc. | Evidence A → |
| Engineering & privacy | 12 / 15 | Responsive neon-green mask brush with undo and before/after toggle; cloud inpainting completes in ~6.5s. | Evidence A → |
Capability · 69/100 · 70%
| Criterion | Score | Rationale | Basis |
|---|---|---|---|
| Task quality | 43 / 65 | Painting over the three test signs and clicking Erase completely removed all RESERVED, SALE 50%, and FREE WIFI text and boards (inpaint_psnr_db = 12.25 dB), though the 600×600 free canvas downscaling lowered unmasked background fidelity to unmasked_psnr_db = 24.89 dB (43.0/65). | Evidence A → |
| Deliverable specifications | 14 / 20 | Free tier processes and exports at 600×600 (8.58% pixel area retained); full 2048×2048 resolution requires Pro (14.0/20). | Evidence A → |
| Processing speed | 12 / 15 | Cloud GPU LaMa / Diffusion object removal completed in ~6.5s (12.0/15). | Evidence A → |
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
- 2026-10-16
Initial real-machine benchmark using Mode A synthetic occlusion Ground-Truth anchor (anchor_inpaint_v1).