Qwen 3.6: what 94 developers actually say.
94 Hacker News opinions from 80 developers, Nov 2025 – Nov 2026. 79 come from people who say they use it. Here is what they praise, what they complain about, and who switched.
The case for Qwen 3.6, in one paragraph.
Qwen 3.6 earns 76% approval across 94 opinions. Developers praise its code quality. In the last 90 days it sits at 71%.
Pick it if you
- Want local inference for privacy-sensitive work
- Value fast answers on a Mac
- Need a low-cost local model for simple tasks
Look elsewhere if you
- Need reliable agentic coding on large repos
- Plan to run quantized local builds for agent tasks
- Need a local model for moderately complex coding
Where it wins and loses.
Mentions across 94 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 | 25 | 6 | 31 |
| Privacy & local use | 21 | 2 | 23 |
| Speed | 13 | 1 | 14 |
How opinion has moved.
Monthly approval, Apr 2026 – Sep 2026. Bars show opinion volume; low-volume months are still included but faded.
View data table
| Month | Approval | Opinions |
|---|---|---|
| Apr 2026 | 82% | 14 |
| May 2026 | 79% | 21 |
| Jun 2026 | 72% | 37 |
| Jul 2026 | 90% | 5 |
| Aug 2026 | 73% | 11 |
| Sep 2026 | 88% | 4 |
What developers say.
Summaries are paraphrased. Each links to the original Hacker News comment.
Praise
4 selectedUses local Qwen 3.6 35B for speed on a 128 GB MacBook.
Finds local Qwen 3.6 fast on a MacBook.
Runs a local Qwen 3.6 MoE on an M3 Max at about 90 tokens/sec to generate tested code.
Used Qwen 3.6 for research and orchestration, praising its speed and capability for the project.
Complaints
4 selectedQwen 3.6 misdiagnosed bugs, gave wrong fixes, and wrote very slow code on the same codebase.
Quantized Qwen 3.6 is not good at agentic tasks and often invents details about documents.
Ran Qwen 3.6 directly on a code review in a 30M+ line repo and exhausted its context window.
Qwen 3.6 35B MoE stalls on bigger projects and struggles to hold large scopes locally.
Questions developers ask.
Is Qwen 3.6 worth it?
Is Qwen 3.6 better than Gemma 4?
Is Qwen 3.6 good for coding?
The data behind this page.
Developers give Qwen 3.6 76% approval across 94 opinions from 80 authors, with 79 firsthand reports. Approval over the last 90 days is 71%, below the overall figure. Monthly samples are uneven: 82% in April, 72% in June on 37 opinions, and 88% in September; the two October opinions scored 25%, too small a sample to read as a trend. Code quality (25 positive, 6 negative), local privacy (21 positive, 2 negative), and speed (13 positive, 1 negative) drive the positive case; reliability is split at 4 and 4.
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.