Claude · Last 7 days
Claude Opus 5
0–100 · higher means more positive community experience, not benchmark performance.
Updated 2026-10-02 14:32 UTC · 2026-09-25 – 2026-10-02 UTCRECENT EXPERIENCE
How the Experience Index is changing
30 days of rolling 7-day scores at each daily snapshot. The vertical scale adapts to the data. Missing or insufficient samples leave gaps; today is still updating.
Daily readings & sample sizes
| Date | Score | n | Scoring window (UTC) |
|---|---|---|---|
| 2026-10-02 | 39.2 | 6 | 2026-09-25T14:32:03+00:00 – 2026-10-02T14:32:03+00:00 |
| 2026-10-01 | 39.2 | 6 | 2026-09-24T23:57:02+00:00 – 2026-10-01T23:57:02+00:00 |
| 2026-09-30 | 39.2 | 6 | 2026-09-23T23:57:03+00:00 – 2026-09-30T23:57:03+00:00 |
| 2026-09-29 | 34.4 | 8 | 2026-09-22T23:57:03+00:00 – 2026-09-29T23:57:03+00:00 |
| 2026-09-28 | 19.5 | 23 | 2026-09-21T23:57:03+00:00 – 2026-09-28T23:57:03+00:00 |
| 2026-09-27 | 20.4 | 26 | 2026-09-20T23:57:03+00:00 – 2026-09-27T23:57:03+00:00 |
| 2026-09-26 | 20.4 | 26 | 2026-09-19T23:57:03+00:00 – 2026-09-26T23:57:03+00:00 |
| 2026-09-25 | 21.1 | 25 | 2026-09-18T23:57:03+00:00 – 2026-09-25T23:57:03+00:00 |
| 2026-09-24 | 21.1 | 25 | 2026-09-17T23:57:03+00:00 – 2026-09-24T23:57:03+00:00 |
| 2026-09-23 | 21.9 | 24 | 2026-09-16T23:57:03+00:00 – 2026-09-23T23:57:03+00:00 |
| 2026-09-22 | 25.0 | 20 | 2026-09-15T23:57:03+00:00 – 2026-09-22T23:57:03+00:00 |
| 2026-09-21 | 38.7 | 6 | 2026-09-14T23:57:03+00:00 – 2026-09-21T23:57:03+00:00 |
| 2026-09-20 | — | 3 | 2026-09-13T23:57:03+00:00 – 2026-09-20T23:57:03+00:00 |
| 2026-09-19 | — | 3 | 2026-09-12T23:57:03+00:00 – 2026-09-19T23:57:03+00:00 |
| 2026-09-18 | 39.5 | 8 | 2026-09-11T23:57:02+00:00 – 2026-09-18T23:57:02+00:00 |
| 2026-09-17 | 37.8 | 10 | 2026-09-10T23:57:03+00:00 – 2026-09-17T23:57:03+00:00 |
| 2026-09-16 | 35.3 | 12 | 2026-09-09T23:57:03+00:00 – 2026-09-16T23:57:03+00:00 |
| 2026-09-15 | 35.4 | 12 | 2026-09-08T23:57:03+00:00 – 2026-09-15T23:57:03+00:00 |
| 2026-09-14 | 36.6 | 13 | 2026-09-07T23:57:03+00:00 – 2026-09-14T23:57:03+00:00 |
| 2026-09-13 | 33.7 | 12 | 2026-09-06T23:57:03+00:00 – 2026-09-13T23:57:03+00:00 |
| 2026-09-12 | 36.6 | 13 | 2026-09-05T23:57:03+00:00 – 2026-09-12T23:57:03+00:00 |
| 2026-09-11 | 40.5 | 9 | 2026-09-04T23:00:04+00:00 – 2026-09-11T23:00:04+00:00 |
| 2026-09-10 | 34.8 | 11 | 2026-09-03T23:00:05+00:00 – 2026-09-10T23:00:05+00:00 |
| 2026-09-09 | 44.7 | 16 | 2026-09-02T23:00:04+00:00 – 2026-09-09T23:00:04+00:00 |
| 2026-09-08 | 47.1 | 19 | 2026-09-01T23:00:03+00:00 – 2026-09-08T23:00:03+00:00 |
| 2026-09-07 | 41.3 | 23 | 2026-08-31T23:00:03+00:00 – 2026-09-07T23:00:03+00:00 |
| 2026-09-06 | 41.5 | 24 | 2026-08-30T23:00:02+00:00 – 2026-09-06T23:00:02+00:00 |
| 2026-09-05 | 39.5 | 23 | 2026-08-29T23:00:03+00:00 – 2026-09-05T23:00:03+00:00 |
| 2026-09-04 | 40.6 | 22 | 2026-08-28T23:00:03+00:00 – 2026-09-04T23:00:03+00:00 |
| 2026-09-03 | 42.3 | 22 | 2026-08-27T23:00:03+00:00 – 2026-09-03T23:00:03+00:00 |
Hacker News · 39.2 · 6 scored comments · Read platform reviews →
ONE MODEL, DIFFERENT COMMUNITIES
Across the communities
Scored separately for each platform, with different samples and audiences. Platform scores are not simply averaged; scored counts exclude quota-only comments.
Scores range from 0 to 100. Select a platform for its trend and reviews. Latest opinion time is not crawler health.
What people think
One user fed both models a script meant to print the max of two numbers; Opus 5 deduced it correctly, while Opus 5.5 wrongly concluded it prints 1/0. One user reported that Claude Opus 5 broke ten things every turn while producing incomprehensible babble during coding, prompting them to switch from Claude Code to Codex.
About the sample and scores
As of 2026-10-02 14:32:03 UTC, Claude Opus 5 in the Claude family has 75 explicitly attributed community comments in the last 7 days; its community score is 20.0/100. Available category scores include General text 18.1/100 (n=45); Reasoning 36.8/100 (n=16); Speed & latency 32.1/100 (n=7). Scoring method and sources
Community reviews
Selected comments from the last 7 days. The balance of excerpts does not represent the share of positive reviews.
2 selected excerpts
Community reports about limits cannot establish official allowances.
Comment context
But right around the time my monthly limit reset this evening, I noticed an extreme shift in communication style and coding behavior that feels suspiciously like Opus 5 (O5). For the last week, O5.5 would take my requirements and immediately get to work, usually knocking out a feature in minutes and doing an incredible job across implementation, UX, architecture, and token efficiency. Tonight, I’m seeing something very different. The first tell was the communication style. O5.5 suddenly started responding to feature requests with these glazing pre-implementation narratives and generic estimates like, “That’s a great idea. That’s two days’ worth of work and it will be amazing,” and then proceeding to implement the feature in the most slopcode-slathered way possible. That was a very specific and particularly annoying characteristic of O5, so I clocked it immediately. The next tell: a sudden reversion to walls-of-text responses when prodded for details on...*anything*.
Comment context
I feed it a script which takes two numbers and prints the max of the two numbers. Opus 5.5 thinks it prints 1/0 instead of the max numbers.
Show original
opus5太慢了
Comment context
Actually I thought Opus 5 was great, and don't feel like 5.5 is some incredible leap forward from it
Comment context
Tried to use muse for PR review and opus 5 was hating on the feedback and instead opus found a different real issue that muse didn't.
Comment context
He was able to make 1 model rocket animation at 4x the speed of regular astra, 1 website redesign and I think 1 more thing but I cant remember.
No selected comments for this filter in the last 7 days.