GLM 5: what 30 developers actually say.
30 Hacker News opinions from 30 developers, Nov 2025 – Nov 2026. 25 come from people who say they use it. Here is what they praise, what they complain about, and who switched.
The case for GLM 5, in one paragraph.
GLM 5 earns 78% approval across 30 opinions. Developers are divided on it.
Pick it if you
- Want strong code quality for the price
- Run agent or app-scripting tasks
- Need more usage for the money
Look elsewhere if you
- Need Opus-level code on demanding tasks
- Need steady reliability at peak hours
- Rely on prompt caching to control costs
Where it wins and loses.
Mentions across 30 opinions. Bars are counts, not votes; a tool can be both praised and criticized on the same topic.
View data table
| Topic | Praise | Complaints | Mentions |
|---|---|---|---|
| Code quality | 9 | 1 | 10 |
How opinion has moved.
Monthly approval, Feb 2026 – Apr 2026. Bars show opinion volume; low-volume months are still included but faded.
View data table
| Month | Approval | Opinions |
|---|---|---|
| Feb 2026 | 90% | 10 |
| Mar 2026 | 70% | 5 |
| Apr 2026 | 70% | 10 |
What developers say.
Summaries are paraphrased. Each links to the original Hacker News comment.
Praise
4 selectedGLM 5.x can produce Emacs org-mode documents for reproducible research.
GLM 5 has been a smooth ride for running their OpenClaw setup.
GLM 5 beat DeepSeek V4 and Kimi K2.6 at app-scripting agent tasks in their tests.
GLM 5 is really good for the price and gives more usage, though peak hours degrade.
Complaints
4 selectedGLM 5.x costs much more for them because caching works poorly.
Tried GLM 5 and judged its code well behind Opus 4.6.
GLM 5 forgot tool-use details and wasted tokens on duplicate-code fixes.
GLM 5 via OpenCode returned gibberish and was very dumb compared with another provider.
Questions developers ask.
Is GLM 5 worth it?
Is GLM 5 better than Claude Opus 4.6?
Is GLM 5 good for coding?
The data behind this page.
GLM 5 has 78% approval across 30 opinions, 25 of them firsthand, with mixed views counted half. Developers like its code quality (9 positive, 1 negative) and price (5 positive, 1 negative). Reliability draws more doubt (4 positive, 2 negative), including peak-hour slowdowns and one bad provider result. Monthly approval was 90% in February and 70% in March and April; samples from May onward are one opinion each, so there is no reliable recent trend.
Every number comes from public Hacker News comments collected through the HN Search API. A language model tagged each comment for sentiment, first-hand use, topic and switches; a human spot-checks samples. No vendor can pay for placement.
Get the monthly report
Which AI coding tools gained or lost developer approval on Hacker News, and why people switched. One email a month.