What commenters asked our summary
In our own words. Each line links to the comment it is based on.
- Whether the individual benchmarks for the Qwen coding model can be published, especially math-related ones.From this comment
- If repeating the same layer block more than twice would yield even greater performance.From this comment
- Why test on an older model version when newer versions have been available for some time.From this comment
- Whether this improvement stems from the same mechanism as the trick of repeating the prompt.From this comment
What they doubted or pushed back on our summary
- They doubt the method will hold up on a balanced set of tasks, even with fine-tuning.From this comment
- If performance drops on other tests, the gain is not very useful.From this comment
- The writing reads as machine-generated, and layer duplication has been tried before.From this comment
- The explanation does not convince because models were not trained to make arbitrary layer outputs feed the next layer.From this comment
What the maker answered our summary
6 replies from the maker in the thread; the points below are the ones we could tie to a comment.
- Said the results are in the repository and listed the exact benchmark task configuration used.From this comment
- Tried repeating the circuit four times, which scored lower though the model still seemed coherent.From this comment
- A patch could change the flow without extra memory, possibly leading to a mixed architecture.From this comment
- Admitted using a language model for drafting, claimed the experiments were original, and said the new part is the sweep tool plus standard benchmark validation.From this comment
What a builder can take from it: A builder should validate layer-duplication gains on a broad benchmark suite and publish all results, since selective improvements may not generalize. our reading
The title, taken apart our calculation
This post got 265 points; the median for all 1442 posts is 112. In AI and agents (352 posts) the median is 108 points and 49 comments.
When it was posted our calculation
Wednesday at 21:00 UTC. Of 1,956 Show HN posts submitted in that UTC hour over the window (2025-11-20 to 2026-11-20), 55 (2.8%) reached 50 points; on Wednesdays it was 2.6% of 8,118. The average over all hours is 2.9%. One post's time does not explain its result; the full table of hours is on the Show HN hub.
Check your title our calculation
Type the title you plan to post. It is compared with 1,442 Show HN posts that reached 50 or more points: its length, its shape, and how posts of that length and shape did. Nothing is sent anywhere.
Type a title above to see where it sits.
Draft my Show HN post
Paste your site address and get title options and a first comment for your own Show HN post, written from what your site says and checked against the numbers above. Nothing is posted for you.
Free to try and no account needed. FoxPlug is for people who ship several projects and keep stalling on telling anyone: it turns what you ship into posts you approve before anything goes out. See what each plan includes.
More AI and agents teardowns
Points, comments, titles and times are Hacker News's own numbers, read from the post itself, last updated Nov 2026. Summaries of what people said are ours, written in our own words, and each one links to the comment it came from. Percentages and medians marked "our calculation" are computed from the posts counted on the page: they describe what happened, not what will happen to your post. FoxPlug posts nothing for you. Plans and limits are on the pricing page. Questions: support.
After the launch post: a changelog your users read
A launch brings visitors once. FoxPlug turns what you ship afterwards into a public changelog and ready posts, so the next visitor finds something new.
Start your changelog See a SaaS changelog example
Free to start. Pro is a month.
FoxPlug Pro
- A public changelog page and feed for each project
- Visual cards and walkthrough videos made from what you shipped
- Posts for X, LinkedIn and dev.to written from your changes
- Connect GitHub and the tools you build with