The Standings → GPT‑5.6 Luna (high)
OpenAI
high reasoning effort · via Codex CLI · judged by Claude Sonnet 5 · not the setting shown on the standings; also run at xhigh50.3 medium49.9
Results from run ‘first’: ten items, ten verdicts.
GPT‑5.6 Luna (high) 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 momentum, relatively weak on the author’s voice. On the mechanical checks it kept to the brief’s word range on 5 of 10 items.
The record: 10 items · 10 verdicts · 0 errors · 2,582 words on average · est. generation cost $0.10
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 The closing telegram is a genuinely gripping, memorable swerve that makes the reader want to know what happened immediately.
Worst The final twist has Jonathan travel from a Budapest sickbed, too weak to travel, to Exeter and total denial of his fiancée within three days, which reads as a shock effect the story hasn't earned rather than a believable turn.
Best It builds a nice bit of unsettling tension near the end when the buried object makes a strange sound and shifts slightly, without ever showing anything coming out of it.
Worst The characters talk back and forth too much like people in a modern conversation, which makes it feel less like the reflective old-fashioned narrator of the original book.
Best The swimming scene is clear and physical, with a real sense of fear turning into a small private triumph.
Worst It keeps stopping to tell you exactly what Edna feels instead of letting her actions and the scene show it.
Best The chapter is easy to follow and hits all four required plot beats clearly and in a sensible order, giving a strong sense of the house's size and gloom on arrival.
Worst The back-and-forth of Mary repeatedly asking 'What did you say?' to the manservant is dragged out so long that it turns into a language lesson and makes Mary sound unusually chatty and inquisitive rather than the guarded, tight-lipped girl established earlier in the book.
Best The confrontation over the word 'tenant' is staged clearly and builds to a sharp, satisfying outburst from Sir Walter.
Worst The scene plays out like a script of quick back-and-forth lines with almost no narrator's commentary or irony, so it doesn't sound like Austen guiding the reader through what people really mean.
Best The teasing back-and-forth between Eve and Psmith at the tea-shop, with his flat repeated denials, has a nicely sustained rhythm of deflection.
Worst The job-interview scene, the chapter's one required set-piece, never actually happens on the page — it's only summarized afterward in dialogue, so the comic potential is thrown away.
Best The chauffeur's blunt contradiction of Alfred's alibi about the motor-car is clean, clear, and easy to follow.
Worst The dialogue is chopped into very short back-and-forth lines that read more like a modern screenplay than Christie's flowing period narration, and it flatly contradicts how the bedroom door was actually opened in the source chapter.
Best The relentless back-and-forth questioning of the widow does build a real, mounting sense of suspicion and pressure.
Worst Almost the entire chapter is short question-and-answer dialogue with almost no description or physical action, so it reads like an interrogation transcript rather than a scene in a novel.
Best The clipped, rapid-fire questioning between Betteredge and Franklin gives the packet handover a genuine, nervy tension.
Worst It drops the thread that the earlier chapter set up so carefully — the boy's prophecy and the identification of 'It' as the Moonstone never comes up, so the letter and packet feel like they belong to a different, blander story about surveillance rather than a cursed diamond.
Best The scene where Jake goes back for Georgette has some real personality to it, and using the correct street name and her correct surname shows careful attention to the source material.
Worst Almost the entire chapter is built from very short one- or two-word question-and-answer lines with almost no physical action or description holding them together, so the conversations feel like a mechanical drill instead of two people talking.
Headline, compliance and craft for each chapter, 0–100.
| Item | Headline | Compliance | Craft | Words |
|---|---|---|---|---|
| Dracula item 015 · after ch. 5 · Bram Stoker | 63.4 | 95 | 67 | 2,627 |
| The War of the Worlds item 006 · after ch. 1 · H. G. Wells | 57.8 | 95 | 61 | 3,045 |
| The Awakening item 016 · after ch. 7 · Kate Chopin | 57.5 | 100 | 57 | 1,781 |
| The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett | 55.7 | 90 | 62 | 2,991 |
| Persuasion item 001 · after ch. 1 · Jane Austen | 51.4 | 95 | 54 | 3,180 |
| Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse | 45.4 | 95 | 48 | 2,534 |
| The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie | 42.8 | 85 | 50 | 2,031 |
| Red Harvest item 023 · after ch. 2 · Dashiell Hammett | 40.8 | 95 | 43 | 1,986 |
| The Moonstone item 011 · after ch. 8 · Wilkie Collins | 40.7 | 90 | 45 | 3,084 |
| The Sun Also Rises item 021 · after ch. 3 · Ernest Hemingway | 28.6 | 80 | 36 | 2,558 |
Words written per item, against each brief’s maximum.
Word count per item vs. brief maximum · run ‘first’
Five of ten items landed inside their brief’s word range. Dracula and The War of the Worlds and The Secret Garden and Persuasion and The Moonstone ran over by 27 and 445 and 591 and 580 and 284 words respectively.
Each verdict names the chapter’s single worst flaw, with evidence. Quoted verbatim from run ‘first’.
It drops the thread that the earlier chapter set up so carefully — the boy's prophecy and the identification of 'It' as the Moonstone never comes up, so the letter and packet feel like they belong to a different, blander story about surveillance rather than a cursed diamond.
The back-and-forth of Mary repeatedly asking 'What did you say?' to the manservant is dragged out so long that it turns into a language lesson and makes Mary sound unusually chatty and inquisitive rather than the guarded, tight-lipped girl established earlier in the book.
Almost the entire chapter is built from very short one- or two-word question-and-answer lines with almost no physical action or description holding them together, so the conversations feel like a mechanical drill instead of two people talking.
The dialogue is chopped into very short back-and-forth lines that read more like a modern screenplay than Christie's flowing period narration, and it flatly contradicts how the bedroom door was actually opened in the source chapter.
Automated screening results across the ten outputs.