Analyze your relationship texts completely free, no card required. Upload a conversation export, confirm both participants and compare reply timing, effort, conflict and repair. Inspect your real preview before choosing an optional full report.
Start with a completely free analysis of your entire text history, with no card required. Your free preview includes 4 evidence-linked findings. An optional full report is $17.99 USD once: your complete relationship story across the full timeline, cinematic narrative, Memory Lane, evidence, metrics, custom charts, and available results from Grey Mirror's 30-analyzer inventory. Grey Mirror Plus is $9.99/month for 1 full report credit each paid month. Unused credits carry forward; keep unlocked reports permanently.
What do I need to upload?
Choose one conversation you have permission to analyze. Use the platform-specific export guides below if you still need the file.
Keep both senders, timestamps and every relevant message-file part. Check that the first and last dates cover the period you want to understand.
Upload the supported export and confirm which participant is you. Check the imported count and dates before interpreting the report.
A relationship text analyzer measures patterns in an exported conversation: who starts contact, how replies change and what follows conflict. Grey Mirror compares the supplied history across both participants and dates, with supporting messages and stated limits. Start with completely free analysis, no card required. The full report is an optional $17.99 one-time unlock or an included Grey Mirror Plus report credit. Text patterns cannot establish private intentions.
Supported exports include iMessage, SMS, WhatsApp, Instagram DMs and Telegram. Keep both people, timestamps and every relevant file part.
Message counts and recorded reply intervals are calculations; labels such as repair or conflict are interpretations to check against the surrounding exchange.
Inspect the imported count and date range before interpreting a chart. Unavailable measurements stay unavailable.
Use the report to prepare a clearer conversation. It cannot establish love, diagnose a person or predict a relationship outcome.
Start with your question
What can you learn from your relationship texts?
Choose one question and inspect the measure that answers it. Several signals can support a pattern; no single count settles what another person feels.
A report you can inspect
Try this reading task in the real sample report
The public Jack and Diane report covers 47,693 messages across 160 days. It is a real report shared with permission; participant names were changed. Use it to inspect the output before supplying your own private history.
Start with a reply-time or initiation chart and check which participant and date range it describes. Then read the written finding beside it and open the available evidence. Ask whether the messages support that interpretation and what offline context could change it.
This separates three different checks: the export was imported correctly, a measurement was calculated from that record, and an interpretation fits the supporting exchange. A convincing narrative alone does not establish all three.
Psychology-informed relationship analysis, with metrics and supporting messages you can examine. Grey Mirror is developed by Wentropy Labs and available through JustLayMe, the consumer web platform Wentropy Labs operates.
Deep analysis, with the inventory to back it up
Grey Mirror has 30 analyzers built on up to 47 models. Full-history analysis compares both participants across communication patterns, with supporting messages and stated limitations.
Follow patterns across the available message history: timing, effort, repair, affection and turning points.
Inspect metric drivers, timeline windows, evidence references and uncertainty in the complete report.
Compare the published analyzer inventory with ScanMyLove and Lucen in our dated comparison guide.
For couples
Understand the conversation you share
Compare how each of you starts contact, responds and attempts repair. Read the same dated evidence together, with room for context the messages cannot show.
Yes. Upload the conversation around the breakup, including the lead-up and any later contact. Grey Mirror can compare changes in initiation, reply gaps, conflict and repair, with dates and supporting messages. It cannot establish why someone left, what they secretly feel, or whether you should get back together.
How to analyze text messages once you have the whole export.
Text message analysis is a sequence, not a verdict. Each step below is something you can check yourself, and each one is what the report does before it says anything about the relationship.
Use the supported export format for your app and keep sender names, timestamps and every relevant file part. Check the available date range: a platform export may omit older, deleted or disappearing messages. Each guide below explains what to upload and how to check it.
Why should an AI text message analyzer read the whole thread?
A screenshot is fast, but it is context-poor. Full-thread relationship analysis needs the full thread to show whether a message is rare, repeated, repaired, ignored, contradicted later, or part of a larger cycle.
A relationship text analyzer measures repair, response timing, effort balance, power, and future planning.
Grey Mirror measures relationship signal families that only become meaningful over time: repair, response timing, effort, positivity, power, planning, communication shifts, emotional momentum, recurring loops, and evidence confidence.
The report is built from the conversation record, not from a guess about the people in it.
Grey Mirror builds the report from the complete conversation record, not from a free-form guess about a few dramatic lines.
Export the conversation from the source platform when possible.
Upload the file through the Grey Mirror flow.
Normalize messages into sender, timestamp, body, and sequence fields.
Resolve participants when labels are ambiguous.
Segment the real timeline into countable windows.
Run route-aware relationship metric families.
Attach evidence, confidence, and evidence-quality states when the payload supports them.
Generate the dashboard, report, and report surfaces.
Evidence, not vibes
Every finding is tied to the messages, the time window, and the confidence behind it.
A useful relationship report makes claims that can be inspected. Grey Mirror is designed around evidence windows, confidence labels, and graceful degradation when the data is missing, ambiguous, or too thin.
The report should explain which metric families moved the score, which participant patterns matter, and whether the signal is strong, partial, or weak.
Show evidence
Evidence may appear as message references, windows, counts, or report-level explanations depending on the user access level and the stored payload.
Insufficient evidence
When the parser needs more messages, timestamps, participant confidence, or repeated examples, Grey Mirror should mark confidence clearly instead of manufacturing certainty.
Report preview
A full-thread report gives you a score, repair and timing breakdowns, a trajectory, and the evidence under each claim.
A Grey Mirror report is designed to combine a high-level read with expandable signal evidence: score, repair, timing, power, future planning, trajectory, evidence areas, and a grounded action plan when supported.
Relationship score and signal summary.
Repair and rupture analysis.
Response timing, silence, and pacing patterns.
Effort imbalance and positivity reciprocity.
Power dynamics and boundary-pressure signals when supported.
Future planning and communication shift trends.
Report and dashboard views for the same run.
Longitudinal change surfaces when later linked runs are available.
How do romantic, platonic, and family routes differ?
The route changes the interpretive lens, not the uploaded text. Romantic, platonic, and family conversations can share timing, repair, and reciprocity mechanics, but the expectations around commitment, care, obligation, and boundaries differ.
Frames reciprocity, emotional labor, reliability, mutual support, and friend-group context more carefully.
Family
Treats obligation, generational roles, boundary pressure, and repeated conflict loops as central context.
Evidence context
Sensitive findings carry confidence labels and evidence windows, never a diagnosis.
Grey Mirror organizes text evidence with confidence labels, evidence windows, and trust-page resources for sensitive situations.
Keep sensitive claims tied to evidence windows and confidence labels.
Use trust-page resources for safety, legal, crisis, or professional-support context.
Read sarcasm, deleted messages, off-platform events, and private context as evidence-quality factors.
It works best with complete exports and correct participant mapping.
Relationship questions, useful measurements and what to check
Your question
Useful evidence
Check before interpreting
Am I carrying the conversation?
Useful evidenceInitiation, questions, follow-ups and reconnection for both participants.
Check before interpretingRead the definition of a new conversation. Several follow-ups in one exchange are not several new starts.
Are their replies getting slower?
Useful evidenceReply intervals across comparable earlier and later dates.
Check before interpretingCheck overnight gaps, work schedules, time zones and missing timestamps. Slower replies alone do not prove lost interest.
Do our arguments actually get resolved?
Useful evidenceDated conflict, acknowledgment and later follow-through, including counterexamples.
Check before interpretingAn apology count is not a repair count. Read what happened after the apology and consider repair that happened offline.
When did our communication change?
Useful evidenceChanges in several measures against this conversation’s earlier baseline.
Check before interpretingOpen messages around the proposed date. A detected change does not establish its cause.
Is affection mutual?
Useful evidenceExpressions of warmth, responses and recurring unanswered bids.
Check before interpretingRead how each person communicates care. A phrase count cannot establish love or include actions outside the thread.
Why should an AI text message analyzer read the whole thread?
Question
Screenshot analyzer
Full-thread Grey Mirror report
Did this happen before?
Screenshot analyzerUsually no timeline depth.
Full-thread Grey Mirror reportChecks recurrence across windows and episodes.
Did repair follow conflict?
Screenshot analyzerOften invisible.
Full-thread Grey Mirror reportMeasures rupture, repair, and aftermath.
Did effort become one-sided?
Screenshot analyzerHard to tell from one moment.
Full-thread Grey Mirror reportCompares initiation, follow-up, planning, and unanswered bids.
Did the relationship cool over time?
Screenshot analyzerUsually too little context.
Full-thread Grey Mirror reportLooks for timing drift, lower reciprocity, and weaker future planning.
How confident is the claim?
Screenshot analyzerRarely explicit.
Full-thread Grey Mirror reportUses evidence and confidence language when supported.
Frequently Asked Questions
What is AI text message analysis?
AI text message analysis uses software to measure communication patterns in an exported conversation: who initiates, how replies change, whether warmth is returned, and what follows conflict. Grey Mirror combines timestamp-based measurements with language-model signals and a written report. The result describes the supplied conversation; it cannot establish someone’s private intentions.
How accurate is an AI text message analyzer?
Accuracy depends on the measurement. Message counts and reply intervals can be checked against an intact export. Detected repair, affection and conflict are model interpretations that can miss sarcasm or offline context. Grey Mirror publishes its methodology, evidence standards and task-specific benchmarks. A confidence score is not an overall accuracy percentage or a probability that a relationship will succeed.
What does Grey Mirror Deep AI Analysis add to the report?
Deep AI Analysis connects the measured findings into a detailed explanation of communication, warmth, conflict, repair and possible next steps. It works from the report data rather than a few selected texts. The complete Jack and Diane sample shows the narrative beside reply-time, repair and timeline charts. Your own Deep AI Analysis and Memory Lane require full-report access.
What makes a text analyzer relationship-aware rather than generic?
A generic sentiment tool scores messages one at a time. A text message analyzer for relationships has to work on the pair: who opened the last twenty conversations, whose replies got slower and when, whether a bid for attention was answered or left hanging, and whether a rupture was followed by repair. None of that is visible in a single message, which is why Grey Mirror reads the full exported history in sequence rather than sampling it. If you want to analyze chat history for relationship patterns instead of collecting isolated verdicts, the whole thread is the minimum unit.
Is there a relationship conversation analyzer that reads the full history?
Yes. Grey Mirror accepts supported WhatsApp, iMessage, Instagram, Telegram and SMS exports and retains the supplied message history for analysis. Check the imported count, first and last dates and participant labels. Individual measurements need sufficient usable evidence, and some analyzers can be unavailable for a particular report. A complete import does not mean every analyzer produced a result; missing messages and off-app context remain outside the record.
How fast are the results?
Grey Mirror shows progress as your export is parsed and analyzed. Processing time depends on the file, message count and queue load; large histories take longer. Once analysis is ready, you can preview your real relationship story. The complete timeline, Memory Lane, charts, metrics, evidence, and available analyzer results have a separate unlock.
What does a relationship text analyzer actually measure?
It measures communication patterns across a message history rather than scoring individual messages: response timing and how it drifts, who initiates, whether conflict is followed by repair, effort balance, warmth that gets returned or ignored, future planning that becomes action, and repeated loops. Each of those only becomes readable once the full sequence is present.
Is this different from sentiment analysis?
Yes. Sentiment analysis scores individual messages as positive or negative. This reads the thread as a sequence: who replies first, how fast, who repairs after conflict, and what changes over months. A message can be positive and still be part of a pattern that is getting worse.
Can AI analyze iMessage conversations?
AI can analyze exported iMessage-style conversation histories when the export preserves enough message text, participants, and timestamps. Grey Mirror is built around full-thread structure instead of isolated screenshots.
How is Grey Mirror different from ChatGPT?
A general AI assistant can summarize or discuss messages you provide. Grey Mirror adds a dedicated workflow for exported conversations: participant mapping, timestamp-based metrics, recurring-pattern analysis, evidence windows, charts and a structured report. Compare whether an answer can show its supporting messages and calculation method, rather than judging only how persuasive it sounds.
Is a screenshot enough to analyze a relationship?
A screenshot can show one moment, while a full thread shows recurrence, timing drift, repair quality, silence gaps, effort imbalance, and whether a dramatic message is normal or unusual for that relationship.
What metrics does Grey Mirror use?
Grey Mirror surfaces metrics such as repair rate, rupture versus repair, response timing, timing drift, turn-taking, effort imbalance, positivity reciprocity, power dynamics, future planning, communication shifts, emotional momentum, recurring loops, and evidence confidence when supported by the payload.
Can Grey Mirror detect emotional distance?
Grey Mirror shows text-based signals that often accompany distance, such as slower replies, fewer warm bids, less future planning, lower reciprocity, and weaker repair.
Can Grey Mirror tell if someone loves me?
No text message analyzer can establish whether someone loves you. Grey Mirror can compare expressions of affection, care, repair, planning and responsiveness in the supplied history. Those patterns can help you prepare for a conversation, but they do not reveal private feelings.
Can Grey Mirror analyze WhatsApp or Instagram DMs?
Grey Mirror is designed around exported message histories. Platform-specific reliability depends on whether the export preserves text, sender, timestamp, and sequence well enough for the parser to normalize safely.
Is Grey Mirror private?
Grey Mirror processes an uploaded export to produce your report; it is not an entirely offline tool. Uploaded conversations are not served as public files. The privacy and deletion page lists processing providers, retention targets and deletion options. Review those terms and consent before uploading a shared conversation.
Can I delete my uploaded conversation?
Yes. Grey Mirror provides authenticated deletion requests for your data and account. The privacy and deletion page explains the scope and signed deletion receipt. Active-system deletion does not mean that mandatory security records or encrypted backups disappear immediately.
Does Grey Mirror train on my messages?
Grey Mirror’s published policy states that uploaded relationship content is not used to train or fine-tune model weights. It is used to generate the requested report and to operate, secure, debug and support the service. External narrative providers receive derived report facts rather than complete raw transcripts or evidence excerpts by default.
What does a relationship health score mean?
It is a structured summary of text-based signal families in a report: timing, repair, reciprocity, warmth, pressure, effort, and trajectory.
How does Grey Mirror handle missing context?
Grey Mirror uses confidence labels and trust-page resources when off-platform events, deleted messages, sarcasm, legal context, or emergency-risk language may affect interpretation.
Is there a text message analyzer app for iPhone?
Yes. The Grey Mirror: Chat Analysis is a free iPhone app on the App Store, and the same analysis runs in any browser, so Android and desktop users can upload from the web. On a Mac, the Grey Mirror Mac app reads your Messages conversations directly, with no export step.
Can it analyze text messages saved as a PDF?
Yes. Upload a PDF of the full conversation as it is, including PDFs made by phone backup tools such as iMazing. The PDF needs to show who sent each message and when; a PDF of a few selected messages has the same gaps as a screenshot.
Is there a free text message analyzer online?
Yes. Start a completely free analysis in your browser, with no credit card required. Upload a supported export, confirm the participants and inspect your real preview. The complete report with the full timeline, Memory Lane, charts, metrics and evidence is an optional $17.99 one-time unlock. A monthly plan is also optional.
Three checks before you trust a relationship finding
Visual guide updated
01
Check the imported history
Compare the message count, first and last dates and participant labels with the conversation you intended to upload.
02
Check the measurement
Read the counting rule and date window. An initiation count, reply interval and detected repair measure different things.
03
Check the interpretation
Open the supporting exchange, look for counterexamples and consider calls or events outside the exported conversation.
Conceptual reading guide. This diagram explains what to inspect; it is not a customer report, diagnostic test or measured relationship score.Use this guide on the question you came with
Use all three checks together: correct coverage, a defined measurement and an interpretation supported by the messages. Start with completely free analysis, no card required.
Chat analysis: what the full conversation adds
Chat analysis means examining the structure and language of a supplied conversation. Basic statistics can describe message counts and recorded timing. A relationship analysis also asks how exchanges develop: whether an emotional bid receives attention, whether an argument becomes more workable, and whether a stated plan is followed. These are different questions, and a feature list should not blur them together.
The export adds the messages you did not choose as highlights. Routine logistics, a return after a pause and an ordinary follow-up can change the reading of a memorable sentence. The important comparison is usually against the conversation’s own earlier behavior. More messages make that comparison possible; they do not guarantee a more certain answer about someone’s feelings.
Inspect three things in a report. First, confirm the participants and the range of history actually processed. Second, read the definition and time window of the measure being used. Third, open the exchanges supporting the interpretation and the limitations attached to it. A compelling animation or a compatibility-style number cannot supply missing context.
MosaicChats publishes a chat-analysis product with advertised sentiment, compatibility and personality dimensions and an AI coach. Grey Mirror organizes the supplied history around relationship timelines, both-person comparisons and supporting exchanges. The dated comparison page links the current official sources. Neither an advertised dimension count nor an analyzer inventory establishes a controlled accuracy winner. Choose according to the question, actual output, privacy terms and access you need.