The Standings → GPT‑6 Luna (medium)
OpenAI
medium reasoning effort · via Codex CLI · judged by Claude Sonnet 5 · not the setting shown on the standings; also run at high50.6 xhigh49.7
Results from run ‘first’: ten items, ten verdicts.
GPT‑6 Luna (medium) scores 50 on craft, 13 points below the field’s 63 and 43 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 the author’s voice. On the mechanical checks it kept to the brief’s word range on 8 of 10 items.
The record: 10 items · 10 verdicts · 0 errors · 1,964 words on average · est. generation cost $0.03
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 50. 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 sharp twist: Jonathan is alive, which should be pure relief, but his odd refusal to accept letters plants a new worry instead of closing the story down.
Worst Several letters, especially Billington's and the run of short telegrams at the end, read like plain modern plot-reporting rather than voices with any real period flavor or personality.
Best The contradiction between Alfred's alibi and Dorcas's account of seeing him come in at half-past one is cleanly set up and genuinely raises suspicion.
Worst The key clue of the whole chapter — the small object Poirot finds — is never described at all, so the scene's big discovery has no substance for the reader to hold onto.
Best It is the cleanest and most legible of the two at hitting every required story point in the right order without any confusion.
Worst It reads like a competent period pastiche rather than Betteredge himself — the jokes, digressions and self-interruptions that make his voice distinctive are mostly missing, and the big emotional beats (like Lady Verinder's prohibition) are stated flatly instead of played out.
Best It moves briskly and clearly through every required plot point without ever losing the thread, so it's easy to follow.
Worst The dialogue and narration sound like a modern writer's paraphrase of Jane Austen rather than the real thing — plain, direct sentences with little of her wit or irony.
Best It ties the interview cleanly back into the Phyllis/Keeble money trouble, so the chapter connects to the wider plot in a sensible way.
Worst Lord Emsworth talks and thinks like a brisk, competent businessman instead of the vague, flower-obsessed forgetful earl the real books make him, so the interview scene doesn't feel like this story.
Best It stays plain and unsentimental throughout, especially in the quiet closing exchange with Georgette at the bal musette.
Worst The dialogue is so clipped and repetitive (long runs of one-word lines) that the scenes feel thin and the people in them barely come alive.
Best The argument between the narrator and Ogilvy over whether the object is 'too regular' to be natural gives the required disagreement real back-and-forth shape.
Worst The narrator states his feelings and conclusions outright instead of dramatizing them, so the story is often told to the reader rather than shown.
Best It hits all the required story points cleanly and in a clear order, so the reader never gets lost.
Worst The writing is plain and generic, so the moor, the house, and the strange sound never feel very vivid or eerie.
Best It's clean and easy to follow, never confusing about what happened or who said what.
Worst The interrogation scenes are just strings of short questions and short denials, so they read like a transcript instead of a real conversation between people.
Best The swim itself is a clean, believable small moment: fear, a few clumsy strokes, and a private laugh of triumph.
Worst Most of the chapter is told in flat, quick summary instead of scenes, so it reads like a list of plot points rather than a piece of the novel, and it spells out its own meaning by having a character repeat the theme almost word for word at the very end.
Headline, compliance and craft for each chapter, 0–100.
| Item | Headline | Compliance | Craft | Words |
|---|---|---|---|---|
| Dracula item 015 · after ch. 5 · Bram Stoker | 59.4 | 95 | 62 | 2,011 |
| The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie | 53.4 | 95 | 56 | 1,993 |
| The Moonstone item 011 · after ch. 8 · Wilkie Collins | 51.7 | 100 | 52 | 2,656 |
| Persuasion item 001 · after ch. 1 · Jane Austen | 51.2 | 100 | 51 | 2,338 |
| Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse | 46.8 | 85 | 55 | 2,774 |
| The Sun Also Rises item 021 · after ch. 3 · Ernest Hemingway | 46.7 | 95 | 49 | 1,098 |
| The War of the Worlds item 006 · after ch. 1 · H. G. Wells | 45.6 | 90 | 51 | 1,911 |
| The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett | 43.6 | 100 | 44 | 1,728 |
| Red Harvest item 023 · after ch. 2 · Dashiell Hammett | 42.3 | 95 | 44 | 1,836 |
| The Awakening item 016 · after ch. 7 · Kate Chopin | 32.3 | 95 | 34 | 1,296 |
Words written per item, against each brief’s maximum.
Word count per item vs. brief maximum · run ‘first’
Eight 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’.
It reads like a competent period pastiche rather than Betteredge himself — the jokes, digressions and self-interruptions that make his voice distinctive are mostly missing, and the big emotional beats (like Lady Verinder's prohibition) are stated flatly instead of played out.
Most of the chapter is told in flat, quick summary instead of scenes, so it reads like a list of plot points rather than a piece of the novel, and it spells out its own meaning by having a character repeat the theme almost word for word at the very end.
Lord Emsworth talks and thinks like a brisk, competent businessman instead of the vague, flower-obsessed forgetful earl the real books make him, so the interview scene doesn't feel like this story.
Several letters, especially Billington's and the run of short telegrams at the end, read like plain modern plot-reporting rather than voices with any real period flavor or personality.
Automated screening results across the ten outputs.