Independent Research

We talk to every companion so you don't have to guess.

Independent AI girlfriend and companion app reviews — hands-on chat trials, privacy and pricing checks, and side-by-side rankings. Most apps promise a companion that remembers you and deliver a chatbot that forgets by the next login. We surface the few that don't.

6 Checks Per App
3 Steps Per Ranking
18+ Adults Only
100% Hands-On Trials
Privacy-First Testing
Hands-On Chat Trials
Transparent Pricing Checks
Section 01

What we look for

Six checks that decide whether an AI companion actually delivers the relationship it advertises — or just trades on a pretty avatar and a free first week.

Criterion 01

Memory and conversation quality

We run the same scripted conversations on every app and come back days later. A companion that forgets your name, your job, or last week's conversation fails the check no matter how good the first chat felt.

Criterion 02

Personality and customization

How far you can shape the companion — appearance, voice, traits, backstory, boundaries — and whether those settings actually change how it talks, or just the picture on the profile.

Criterion 03

Privacy and data handling

What the app stores, whether your chats are used to train models, how you delete an account, and what the privacy policy actually says. Apps with vague or missing answers get flagged.

Criterion 04

Pricing transparency

What the free tier really includes, what sits behind the paywall, how subscriptions renew, and whether credits or tokens quietly meter the features shown on the sales page.

Criterion 05

Voice, photo, and media features

Voice calls, selfies, image and video generation. We test whether each feature works as advertised at the tier it's sold on, not just in the app store screenshots.

Criterion 06

Safety and consent controls

Age gates, content settings, how the companion handles boundaries, and what happens when a user brings up a crisis. Apps that manage this well rank higher; the ones that ignore it get a limitations note.

Section 02

How we evaluate

Every category runs through the same three steps. We start from what a companion app is actually sold on, build the scoring framework, then measure each app against it.

Step 01

Framework first

For each category — AI girlfriends, roleplay companions, voice-first apps — we pin down what the apps promise: memory, personality, voice, privacy, price. Those promises become the scoring criteria before we open a single app.

Step 02

Hands-on trial

We create accounts, run the same scripted conversations on every app, and return over several days to test memory, paywalls, cancellation, and the features on the sales page. Vague privacy policies and hidden credit systems are flagged.

Step 03

User-outcome cross-check

We read app-store reviews, forum threads, and community discussions, sorted by satisfaction, billing complaints, and privacy concerns. The patterns inform the limitations section of each ranking.

About the Team

Independent editorial research

Every review is researched, written, and edited in-house by Théo Lacombe, our lead reviewer.

TL
Lead Reviewer
Editorial Team

Théo Lacombe

Lead Reviewer · Meet After Hour Editorial

Théo leads companion-app research at Meet After Hour. The work centers on hands-on chat trials, privacy-policy analysis, and pricing audits across the AI girlfriend, roleplay, and voice-companion categories. Every ranking on this site is built around the framework Théo maintains: pin down what an app promises, test whether it delivers over days rather than minutes, then hold every app to the same bar.

Read more about Théo