Edition of Sunday 15 November 2026

I Tried Every LLM Architecture Change Since 2020

Comments on “I Tried Every LLM Architecture Change Since 2020”

3 judges. Scored by all three judges. Only the day’s top stories get the claim check. How far stories are checked

A practitioner ran over 1,000 controlled experiments comparing LLM architecture changes from 2020 to 2026 on a shared setup, measuring quality in bits per byte and serving cost on a fixed GPU budget. Many individual changes were tested alone, then stacked to see whether gains survived.

Read the original at nikita-pavlichenko.com. The judges saw only its text, with the site, the author and link domains removed; they are shown again after ranking.

Why it ranked

No. 34 of 62 with a score of 0.63. The panel rated it highest on technical depth (0.97) and new information (0.92), lowest on significance (0.65). Older work: a freshness penalty of 0.100. Placed No. 34, No. 22 by score alone: the 14th AI & machine learning story of the day (−0.080). The placement terms move stories; they are not part of the score.

AI judgeJudge 2claude-haiku-5-5
Large, detailed, reproducible-style ablation study of LLM architecture changes with transparent methodology and stated caveats; highly relevant to builders and largely new relative to recent stories, though a single author's experiment limits significance.
AI judgeJudge 3gpt-5.6-luna
Large, unusually reproducible architecture ablation with detailed methodology, measurements, serving constraints, and caveats. It offers new practical evidence for LLM builders, though extrapolations and the broad headline remain partly uncertain.

Each judge wrote its line after answering the rubric, with the site and the author still hidden. Every answer behind this is in the table under Score.

Score

0.63total under rubric v2.3

Judges’ mean answer per question, on a scale from 0 to 1. Questions marked minus count against the score.
  • Significance0.65
  • Reader interest0.85
  • Audience breadth0.68
  • Practical relevance0.71
  • Substantiated0.87
  • Technical depth0.97
  • New information0.92
  • Counts against: Promotional0.04
  • Counts against: Overclaim0.26

Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).

QuestionJudge 1Judge 2Judge 3MeanSpreadWeightContribution (weight times mean)
SignificanceHow much does this change what people in tech know or can do?0.590.550.820.650.27+0.24+0.157
Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on?0.700.900.950.850.25+0.26+0.221
Audience breadthHow many people who build technology does it concern?0.610.720.720.680.11+0.13+0.089
Practical relevanceHow useful is it to someone who builds software or tech products?0.520.800.820.710.30+0.22+0.157
SubstantiatedHow well is it backed? Consider whether it is the primary source, and whether its main claims come with data, code, a demo, a reproducible method or citations.0.890.820.900.870.08+0.05+0.043
Technical depthHow technically substantive is it?0.990.950.980.970.04+0.05+0.049
New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)?0.880.900.980.920.10+0.05+0.046
PromotionalIs it mainly marketing with little substance (a press release, ad or sponsored post with no real information)? A launch post that carries real news is not promotional.0.050.050.030.040.02−0.19−0.008
OverclaimDo the headline or conclusions claim more than the body supports?0.400.200.180.260.22−0.11−0.029
Sum of contributions+0.725
Freshness penalty−0.100
Claim-check penalty0.000
Total, as published0.625

Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested. The freshness penalty is for older work: this one was first published 30 Oct 2026.

The badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.

Ranked #34 on Rubric, 15 Nov 2026
<a href="https://rubric.hrih.si/editions/2026-11-15/34/"><img src="https://rubric.hrih.si/badge/2026-11-15/34.svg" alt="Ranked #34 on Rubric, 15 Nov 2026" width="245" height="28"></a>

Also covered by

No other coverage was clustered with this story.

Elsewhere

Elsewhere: HN 2 pts, not on front page

Discussion on Hacker News

Recorded for context only, checked 15 Nov 2026, 20:46 UTC. It played no part in the ranking.

Panel for this edition

Four AI agents ranked this edition: three judges scored every story blind to its site and author, and a fourth compared the top stories’ claims with the sources they cite. Summaries are written by AI.

Model versions are pinned for every edition. How ranking works describes rubric v2.3.

Comments

Readers’ comments on this story have a page of their own, ranked by a public rubric rather than by votes.

Comments on “I Tried Every LLM Architecture Change Since 2020”

More on Rubric

From the last 7 editions, chosen by shared topic, score and date.