Mistrial – Re run your LLM judge and see how often it flips
Comments on “Mistrial – Re run your LLM judge and see how often it flips”
3 judges. Scored by all three judges. Only the day’s top stories get the claim check. How far stories are checked
Mistrial is an open-source tool that re-runs an LLM judge across repeated trials, judge models, and rubric rewordings to measure how often its verdicts flip.
Why it ranked
No. 86 of 155 with a score of 0.59. The panel rated it highest on new information (0.91) and practical relevance (0.81), lowest on significance (0.48).
- AI judgeJudge 2claude-haiku-5-5
- Release of a tool for measuring LLM judge reliability, with a detailed README and reproducible demo runs; useful to AI builders, a niche project with heavy self-promotional framing and self-reported stats.
- AI judgeJudge 3gpt-5.6-luna
- A technically detailed launch of a practical LLM-judge reliability CLI, with reproducible commands, metrics, examples, CI integration, and supporting references. It is highly relevant to AI builders, though claims rely largely on author-run demonstrations and the README is somewhat promotional.
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
0.59total under rubric v2.2
- Significance0.48
- Reader interest0.76
- Audience breadth0.54
- Practical relevance0.81
- Substantiated0.76
- Technical depth0.79
- New information0.91
- Counts against: Promotional0.28
- Counts against: Overclaim0.36
Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).
| Question | Judge 1 | Judge 2 | Judge 3 | Mean | Spread | Weight | Contribution (weight times mean) |
|---|---|---|---|---|---|---|---|
| Significance | 0.63 | 0.25 | 0.55 | 0.48 | 0.38 | +0.24 | +0.114 |
| Reader interest | 0.77 | 0.62 | 0.90 | 0.76 | 0.28 | +0.26 | +0.198 |
| Audience breadth | 0.58 | 0.50 | 0.55 | 0.54 | 0.08 | +0.13 | +0.071 |
| Practical relevance | 0.94 | 0.70 | 0.80 | 0.81 | 0.24 | +0.22 | +0.179 |
| Substantiated | 0.96 | 0.60 | 0.72 | 0.76 | 0.36 | +0.05 | +0.038 |
| Technical depth | 0.99 | 0.60 | 0.78 | 0.79 | 0.40 | +0.05 | +0.040 |
| New information | 0.92 | 0.90 | 0.92 | 0.91 | 0.02 | +0.05 | +0.046 |
| Promotional | 0.11 | 0.45 | 0.28 | 0.28 | 0.34 | −0.19 | −0.053 |
| Overclaim | 0.36 | 0.35 | 0.38 | 0.36 | 0.03 | −0.11 | −0.040 |
| Sum of contributions | +0.593 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.593 | ||||||
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 badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.
<a href="https://rubric.hrih.si/editions/2026-10-28/86/"><img src="https://rubric.hrih.si/badge/2026-10-28/86.svg" alt="Ranked #86 on Rubric, 28 Oct 2026" width="245" height="28"></a>Also covered by
No other coverage was clustered with this story.
Elsewhere
Elsewhere: HN 1 pts, not on front page
Recorded for context only, checked 28 Oct 2026, 06:47 UTC. It played no part in the ranking.