Claude · Last 7 days
Claude Sonnet 5
0–100 · higher means more positive community experience, not benchmark performance.
Updated 2026-10-02 15:22 UTC · 2026-09-25 – 2026-10-02 UTCWhat people think
One user finds that Opus 5.5 takes shortcuts on small side tasks, while Sonnet and Fable would have completed them properly. One user found their month-long experience with codex/GPT 5.6 noticeably better than Claude Sonnet 5 overall.
About the sample and scores
As of 2026-10-02 15:22:03 UTC, Claude Sonnet 5 in the Claude family has 30 explicitly attributed community comments in the last 7 days; its community score is 32.5/100. Available category scores include General text 38.4/100 (n=19); Speed & latency 34.9/100 (n=5). Scoring method and sources
RECENT 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.
Tap or use ← → for dates, scores and sample sizes. Blank dates have insufficient data.
No trend available
RedNote · Not enough opinions · 0 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.
Community reviews
Selected comments from the last 7 days. The balance of excerpts does not represent the share of positive reviews.
0 selected excerpts
Community reports about limits cannot establish official allowances.
Comment context
But after a few tasks I realised it was omitting huge details of complex tasks that Fable just would not have missed. It instead creates a version that passes its own tests and asserts that the job is done (which to be fair, all LLMs do). But in comparison to 5.1, its like it takes intentional shortcuts to get the job 'done'. It looks like it's been tuned to trim "non-essential" aspects of a task to get to the end faster. And I don't think this is isolated to large tasks that have room for interpretation. I've caught doing the same with much smaller side tasks that both Sonnet or Fable would have nailed every time.
Show original
啊？我还觉得sonnet5挺好用的，居然扑街了？
Comment context
I built an adversarial esoteric programming language to benchmark LLM models and just ran it on Sonnet 5.5 It does worse than Sonnet 5.
Comment context
At work we have copilot and even with 750$ credits I use opus very sparingly and I even replaced it with sonnet 5 recently because I wasn't very impressed with the results, can't wait we enable opus 5.5 next week.
Comment context
I asked it to find online prices for some industrial automation items.
No selected comments for this filter in the last 7 days.