Editorial standards
Review Methodology
A reproducible explanation of what the scores measure, how category weights work, and when prices and ratings are updated.
Published:
Rating model
Each service receives a 0–10 score in six categories. A score of 5 represents a usable but materially limited result, 7 represents a competitive result with clear trade-offs, 9 represents a leading result supported across most checks, and 10 is reserved for an exceptional result without a meaningful gap in that category.
25%
Chat quality and memory
Context continuity, repetition control, persona stability, message access, and memory claims.
20%
Character customization
Appearance, personality, backstory, scenario controls, editability, and private character options.
20%
Multimedia tools
Documented image, voice, and video capabilities together with their limits and credit costs.
20%
Price and value
Entry price, renewal terms, free access, included usage, and likely extra media spending.
10%
Privacy and transparency
Published privacy language, account controls, visibility settings, and clarity about service limits.
5%
Mobile usability
Availability and practical usability of chat, account, and media controls on a mobile viewport.
Calculation
The published score equals the sum of each category score multiplied by its weight. The result is rounded to one decimal place. Category weights total 100%.
(Chat × 0.25) + (Customization × 0.20) + (Media × 0.20) + (Value × 0.20) + (Privacy × 0.10) + (Mobile × 0.05)
Testing process
The editorial workflow checks the free-access path and billing presentation, uses the same character setup across multiple sessions, edits a persona rule to check whether later replies follow it, observes continuity and repetition, attempts one action for each available image, voice, or video type, and checks whether the relevant controls remain usable on a mobile viewport.
- Record the advertised entry price, renewal structure, free-access limit, and separate credit costs
- Use a consistent character concept so customization depth can be compared
- Revisit a conversation to check continuity claims and persona stability
- Attempt one image, voice, and video action where the feature is available
- Check account, chat, and media controls on desktop and mobile layouts
- Separate confirmed functions from unconfirmed or company-reported claims
Tie-breakers
If weighted totals are equal, the service with more complete supporting evidence ranks first. If evidence completeness is also equal, the lower comparable twelve-month cost ranks first. Alphabetical order is the final tie-breaker.
Price update rules
Prices receive a quick review each month, a complete recheck each quarter, and an unscheduled review when a material plan change becomes known. The displayed date changes only after a meaningful check; it is not generated automatically on every build.
Evidence limitation
The current prices were confirmed by the site owner as accurate for the August 2, 2026 snapshot. The original provider URLs were not retained in this repository, so the source register labels that limitation instead of presenting reconstructed links as historical evidence.