ChatGPT app icon

ChatGPT release health

By OpenAI. Ranked 5th of 7 global AI apps. Scored on US Play Store reviews.

Google PlayApp StoreWebsite

As of 16 October 2026, ChatGPT's Android app scores 70/100 on Quash's Release Health Index (Solid Foundation), ranking 5th of 7 global AI apps. 0.7% of the 173,400 reviews analysed describe the app failing. It ships roughly every 6.8 days. A score for iOS is coming.

Last updated 16 October 2026. Scores are refreshed monthly.

Android

70/ 100

Solid Foundation

0.7% of reviews describe the app failing

1-star share across the last 10 builds. Dashed line is this app's median, 13.1%
  1. v1.2026.160: 13.1% 1-star
  2. v1.2026.125: 12.7% 1-star
  3. v1.2026.188: 7.4% 1-star
  4. v1.2026.111: 2.6% 1-star
  5. v1.2026.139: 15.2% 1-star
  6. v1.2026.104: 20% 1-star
  7. v1.2026.146: 14.9% 1-star
  8. v1.2025.350: 29% 1-star, flagged as a spike
  9. v1.2026.258: 5.5% 1-star
  10. v1.2026.265: 6.7% 1-star
v1.2026.160v1.2026.265

1 build spiked well above the median

Store rating
4.78 ★
Ratings
64,750,441
Reviews analysed
173,400
Ships every
6.8 days
Current version
1.2026.279
Last update
2026-10-11

iOS

Coming soon

iOS scores are on the way. This build is listed on the App Store.

View on the App Store

What ChatGPT's score says

ChatGPT holds 4.78 stars from 64,750,441 ratings across 1,000,000,000+ installs. Release Health measures something the store rating does not: whether recent builds are shipping cleanly. On that, the Android build scores 70/100 — solid foundation. A score for iOS is coming. That places it 5th of the 7 global AI apps we track.

How this score is calculated

Five dimensions, 100 points, all derived from public store data — the reasoning behind each is shown so any score can be retraced. Two carry 35 points each and three carry 10, in proportion to how well each predicted future problems when tested against months of reviews it had never seen.

Severe defect rate
32.7/35

0.4% of reviews report a severe defect

Hard failures
12% of bug reports name a crash or freeze
High-stakes flows
14% of bug reports hit payments, login or data
Areas affected
3 of 3 high-stakes areas
Regression severity
16.6/35

builds that regress do so by 8.2 points on average

Ships that slipped
4 of 9 landed worse than the one before
Worst ship
+14.1pp above the build before it
Regression frequency
6.6/10

22% of ships land 5+ points worse than the one before

Build consistency
4.7/10

26.4 points separate the best and worst build

Defect persistence
8.9/10

at worst 10% of the window ran consecutively above its own defect median

Indented figures describe the defects behind each rate and carry no points. The score is the sum of the five rows.

Recent builds

Build1-starvs median
v1.2026.2652026-09-24
6.7%-6.4pp
v1.2026.2582026-09-20
5.5%-7.6pp
v1.2025.3502026-09-17spike
29%+15.9pp
v1.2026.1462026-09-17
14.9%+1.8pp
v1.2026.1042026-09-17
20%+6.9pp
v1.2026.1392026-09-16
15.2%+2pp
v1.2026.1112026-09-16
2.6%-10.5pp
v1.2026.1882026-09-16
7.4%-5.7pp
v1.2026.1252026-09-16
12.7%-0.4pp
v1.2026.1602026-09-16
13.1%0pp

Each build is measured over its first 14 days, so builds compare like for like. This app's own median is 13.1%. 1 build landed more than 12 points above it.

Method notes

  • Rating trend is derived from timestamped review scores, not the store's aggregate rating.
  • Android release dates are reconstructed from the versions reviewers were running.

ChatGPT update history & release cadence

Ships roughly every 6.8 days, based on 4 releases dated from the review stream. Last update: 2026-10-11.

  1. v1.2026.2792026-10-09
  2. v1.2026.2722026-10-02
  3. v1.2026.2652026-09-24
  4. v1.2026.2582026-09-20

ChatGPT issues — what users are reporting

0.7% of the 173,400 reviews analysed describe the app failing — not working, not opening, freezing or erroring.

  • highPremature or incorrect usage/attachment limit blocks on free tier
  • highMemory and context loss within and across chats

9 defect themes found in ChatGPT across 8 shipped versions.

Every theme, the versions each affects, what your users said, and the 7 tests that would have caught them.

Common questions

Put this score on your site

ChatGPT release health score
<a href="https://quashbugs.com/release-health/chatgpt"><img src="https://quashbugs.com/api/release-health/badge/chatgpt" alt="ChatGPT release health score" width="280" height="54"></a>

ChatGPT vs other AI apps

See what Quash would test first on ChatGPT

Quash turns bugs like these into a QA test plan and runs it on real devices, with no scripts to maintain.

Book a demo