Claude Opus 5 Faces Backlash: Benchmarks Soar, Users Complain of Bloat
gerardsans · x · 2026-08-21
Users report severe usability issues with Claude Opus 5, suggesting the "just scale more" approach may be hitting a wall in user experience.
- Verbose Output: Despite instructions to be concise (e.g., using Zinsser's style), the model produces extremely wordy and frustrating outputs.
- Functional Failures: The model reportedly argues with instructions, stops mid-task, and behaves unusably, leading some users to cancel subscriptions.
- Divergence: While benchmark scores may rise, the daily utility is declining, raising questions about the effectiveness of scaling laws for practical use.
Related event: Claude Opus 5 Users Revolt as Benchmarks Soar but Daily Usability Crumbles(5 posts)→
More from Models
- Monitors Detect Significant Behavior Shift in Claude Opus — altryne · 2026-08-21
- OpenAI pauses largest training run ever as DeepSeek 'cooks again' — Fireship · 2026-08-21
- DeepSeek V4 Pro benchmarks close to Opus 5 on KernelBench-Hard — teortaxesTex · 2026-08-21
- Liquid AI releases DSpark draft models, speeding up decoding by up to 4x — JosephJacks_ · 2026-08-21
- Users report Qwen Uncensored still frequently refuses requests — BLUECOW009 · 2026-08-21
- Pander Score Leaderboard Reveals Sycophancy Differences in Major AI Models — RobbWiller · 2026-08-21