Claude
2,733 opinions tracked this week · 7-day rolling window
30-day experience score
Version breakdown
Only comments that explicitly name a version contribute to its score.
| Claude Opus 5View lifecycle → | 23.3−14.3 | n=206 | Confidence: High | |
|---|---|---|---|---|
| Claude Fable 5View lifecycle → | 57.9+20.3 | n=154 | Confidence: High | |
| Claude Opus 4.8 | 66.7+29.1 | n=17 | Confidence: Medium | |
| Claude Opus 4.6 | 65.8+28.3 | n=16 | Confidence: Medium | |
| Claude Sonnet 5 | 39.3+1.8 | n=16 | Confidence: Medium |
2,522 generic family mentions are included in the family score, not any version score.
By experience dimension
| General text | 28.5 | n=1,308 | |
|---|---|---|---|
| Coding | 54.3 | n=518 | |
| Reasoning | 53.1 | n=503 | |
| Image / vision | 61.9 | n=8 | |
| Video generation | not enough data | n=0 | |
| Roleplay / creative | 57.1 | n=60 | |
| Safety & refusals | 10.0 | n=131 | |
| Speed & latency | 27.1 | n=246 | |
| Local deploy | not enough data | n=3 |
Top positive drivers
- strong reasoning Reasoning ×198
- problem solved General text ×121
- problem solved Coding ×93
- competitor flight General text ×53
- speed quality Speed & latency ×48
Top negative drivers
- ux format General text ×146
- performance issue Speed & latency ×116
- creativity down General text ×93
- safety overreach Safety & refusals ×92
- performance issue General text ×89
What users are saying
Selected from the latest 7 days, prioritizing recency, evidence quality, confidence, and engagement.
🔴 “the "Claudish" has become so painful to read that it just takes me out of whatever task I was working on”
via Hacker News · General text · 2026-08-26
🟢 “From my own tests, the best model for writing complex, nuanced prose is Opus 4.6.”
via Hacker News · Roleplay / creative · 2026-08-26
🔴 “Claude's default writing style tends to be more verbose than GPT's. GPT models are generally more concise when coding, while Claude often explains more.”
via Hacker News · Coding · 2026-08-26
🔴 “It proceeded to point out all the ways it was different and offer to "fix" the one I rewrote to match... the other documents. Which I was telling it to reference for how -not- to write them.”
via Reddit · Reasoning · 2026-08-26
⚪ “OP, claude's tendency to make up words is known as "nominalisation" and it's mostly a way to pack a lot of contextual information into as little space as possible. Probably to save tokens. I've set claude's global instructions to specifical”
via Reddit · General text · 2026-08-26