Claude Opus 5: what developers actually say
We read every Hacker News comment from Nov 2025 – Nov 2026 that gave a real opinion on Claude Opus 5. 136 of the 162 opinions come from people who say they use it themselves.
Developer opinion on Claude Opus 5 is net negative: approval is 33% across 162 opinions from 148 authors, and 30% over the last 90 days. Code quality draws the most discussion and leans negative at 24 positive against 36 negative, while reliability is the sharpest weakness at 1 positive against 26 negative. Approval fell from 46% in July to 33% in August and 17% in September. The October figure of 75% comes from 6 opinions, too small a sample to read as a recovery.
Pick it if
- Want a low cost option for routine coding
- Need quick drafts and scripts
- Will supply tight instructions and skills
Look elsewhere if
- Need dependable output on every task
- Expect it to follow a detailed spec
- Depend on predictable usage limits
Strengths and weaknesses
Each row counts opinions that mentioned the topic. Bars to the right of the line are praise, bars to the left are complaints.
How opinion has moved
What developers say
Praise
Complaints
Summaries are paraphrased. Each links to the original comment.
Claude Opus 5 compared
FAQ
Is Claude Opus 5 worth it?
The data leans negative. Approval is 33% across 162 opinions, and reliability drew 26 negative mentions against 1 positive. Cost is the one aspect with a positive balance, at 4 positive and 3 negative.
Is Claude Opus 5 better than GPT-5.6 Sol?
Developers in the data more often prefer GPT-5.6 Sol, with 7 naming it over Opus 5 while 3 named Opus 5 over it. Claude Opus 4.8 also drew 7 preferences over Opus 5.
Is Claude Opus 5 good for coding?
Code quality is the most discussed aspect and the balance is negative, at 24 positive against 36 negative mentions. Some developers report strong results when they give detailed instructions and skills, while others report avoidable mistakes on every task.