The Standings → GPT‑6 Luna (xhigh)
OpenAI
xhigh reasoning effort · via Codex CLI · judged by Claude Sonnet 5 · not the setting shown on the standings; also run at high50.6 medium47.3
Results from run ‘first’: ten items, ten verdicts.
GPT‑6 Luna (xhigh) scores 52 on craft, 11 points below the field’s 63 and 41 points short of the authors themselves. Its shape differs from the field’s: relatively strong on restraint and folding the brief into the story, relatively weak on period detail and building scenes. On the mechanical checks it kept to the brief’s word range on 6 of 10 items and is unusually free of stock phrases.
The record: 10 items · 10 verdicts · 0 errors · 2,008 words on average · est. generation cost $0.06
Eight dimensions, each 0–100, averaged over ten items; the author’s real chapter was in the room every time.
Craft by dimension, 0–100 · judged by Sonnet 5 beside the author’s real chapter
The authors themselves average 93 on the five writing dimensions, the field 63, this model 52. Hover or focus a bar for the field and the authors on that dimension.
The judge’s one-sentence best quality and worst flaw for each chapter, written for a general reader.
Best It is clear and well organized, hitting every required plot point cleanly with tidy, easy-to-follow dialogue.
Worst Most of the chapter tells us what people think and decide instead of dramatizing it, so it reads more like minutes of a meeting than a scene from a novel.
Best It reliably delivers every required plot beat in order without losing the thread, so the chapter is easy to follow.
Worst Most of the chapter tells the reader what happened and how Mary felt in flat, generic sentences instead of showing it happening, and there's a confusing moment where the housekeeper calls the manservant by the wrong name, Martha, when he's already been named John.
Best It stays genuinely restrained, never over-explaining Jake's feelings or forcing melodrama into the quiet scenes.
Worst The dialogue leans so hard on short one-line question-and-answer exchanges that scenes start to feel like a script exercise rather than people talking, and there's a plain contradiction about whether Jake keeps the envelope or hands it to Georgette.
Best It is tidy and clear-headed: every required plot point happens cleanly and in order, and the reader is never confused about what just occurred.
Worst Psmith talks like a generic clever gentleman rather than himself — he never once calls anyone 'Comrade' or launches into the cascading mock-heroic speeches that are his trademark, so he reads like a placeholder for Psmith rather than Psmith.
Best The moment where the stable-hand Baily quietly catches Alfred in a lie to his face is genuinely tense and well-paced, with restraint that suits the period.
Worst After a strong opening, the investigation collapses into a fast summary of interviews rather than scenes we actually watch happen, and the clue Poirot finds is described so vaguely ('a tiny square of paper... a dark stain') that it never feels like a real object.
Best The quiet closing image — the Op laying Thaler's card next to the telegram receipt in a drawer before the lights go out — is a nicely understated way to end the day on an unresolved note.
Worst The bank scene and the widow's confession never connect to each other, so the chapter feels like two separate interviews stapled together just to check boxes, and the widow's account leaves a hole (she says she never reached the body, yet still has something like blood on her shoe with no explanation).
Best It hits every required plot point cleanly and in clear, easy-to-follow order.
Worst The writing is flat and plain, more like a summary of events than Chopin's rich, wandering interior voice.
Best It cleanly delivers every required story element in warm, readable letters, especially Lucy's cheerful update about the wedding plans.
Worst The chapter tells you what happened (the dinner, the letters, the worry) more than it lets you feel it, so scenes like the dinner read as a summary instead of something lived.
Best It moves through the required events cleanly and clearly, so the reader never loses track of what is happening.
Worst The writing is clean but generic, so the dialogue and descriptions rarely feel like a particular time, place, or person.
Best It checks off the required plot points cleanly, so a reader never loses track of what has actually happened.
Worst It reads like a flat modern summary of events rather than Betteredge talking to us — the jokes, the fussing, the little habits of the real narrator are almost entirely missing.
Headline, compliance and craft for each chapter, 0–100.
| Item | Headline | Compliance | Craft | Words |
|---|---|---|---|---|
| Persuasion item 001 · after ch. 1 · Jane Austen | 58.5 | 100 | 58 | 2,405 |
| The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett | 53.3 | 95 | 56 | 2,178 |
| The Sun Also Rises item 021 · after ch. 3 · Ernest Hemingway | 51.4 | 95 | 54 | 1,557 |
| Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse | 51.3 | 95 | 54 | 2,666 |
| The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie | 51.1 | 95 | 54 | 1,831 |
| Red Harvest item 023 · after ch. 2 · Dashiell Hammett | 48.3 | 95 | 51 | 1,929 |
| The Awakening item 016 · after ch. 7 · Kate Chopin | 48.1 | 100 | 48 | 1,632 |
| Dracula item 015 · after ch. 5 · Bram Stoker | 46.9 | 90 | 52 | 1,741 |
| The War of the Worlds item 006 · after ch. 1 · H. G. Wells | 46.6 | 100 | 47 | 1,761 |
| The Moonstone item 011 · after ch. 8 · Wilkie Collins | 41.7 | 95 | 44 | 2,380 |
Words written per item, against each brief’s maximum.
Word count per item vs. brief maximum · run ‘first’
Six of ten items landed inside their brief’s word range.
Each verdict names the chapter’s single worst flaw, with evidence. Quoted verbatim from run ‘first’.
The bank scene and the widow's confession never connect to each other, so the chapter feels like two separate interviews stapled together just to check boxes, and the widow's account leaves a hole (she says she never reached the body, yet still has something like blood on her shoe with no explanation).
Most of the chapter tells the reader what happened and how Mary felt in flat, generic sentences instead of showing it happening, and there's a confusing moment where the housekeeper calls the manservant by the wrong name, Martha, when he's already been named John.
After a strong opening, the investigation collapses into a fast summary of interviews rather than scenes we actually watch happen, and the clue Poirot finds is described so vaguely ('a tiny square of paper... a dark stain') that it never feels like a real object.
The dialogue leans so hard on short one-line question-and-answer exchanges that scenes start to feel like a script exercise rather than people talking, and there's a plain contradiction about whether Jake keeps the envelope or hands it to Georgette.
Automated screening results across the ten outputs.