The Standings → GPT‑5.6 Luna (xhigh)
OpenAI
xhigh reasoning effort · via Codex CLI · judged by Claude Sonnet 5 · the setting shown on the standings; also run at medium49.9 high48.4
Results from run ‘first’: ten items, ten verdicts.
GPT‑5.6 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 momentum, relatively weak on folding the brief into the story. On the mechanical checks it kept to the brief’s word range on 4 of 10 items.
The record: 10 items · 10 verdicts · 0 errors · 2,685 words on average · est. generation cost $0.12
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 chaotic interview scene, with Lord Emsworth wandering off-topic and losing his own thread until Baxter bursts in to rescue it, is a fun, true-to-character bit of comedy.
Worst Much of the dialogue is built from very short back-and-forth lines that read more like a modern screenplay volley than the winding, digressive sentences this author is known for.
Best The mystery-sound ending lands cleanly and Mary's stubborn, questioning nature comes through clearly and consistently.
Worst Much of the dialogue, especially Craven's station scene and Mrs. Medlock's houserules, is stated flatly as information rather than played out as a lived moment, so it reads more like a summary of events than a scene.
Best The family conference is cleanly staged and easy to follow, with each required event landing clearly and the shift when Elizabeth backs Anne registering as a real turn.
Worst The characters mostly just state their positions in plain arguments back and forth, so it reads more like a transcript of a meeting than Jane Austen's layered, ironic narration.
Best The swimming scene builds a believable arc from fear to a quiet, private triumph without overstating Edna's feelings.
Worst The dialogue, especially the lady in black's lines and much of Adele's, sounds like plot information being delivered rather than people actually talking.
Best It picks up the story in the exact instant the first chapter left off, so the handoff between chapters feels seamless.
Worst It gives away too much too soon — a seam that seems to be unscrewing itself and a character flatly saying 'there is something alive in it' — which drains the mystery instead of building it.
Best The vanished-check mystery is tracked with real procedural logic (stub number, timing, whether it could have been cashed elsewhere), so the puzzle actually gets more precise rather than just restated.
Worst Three separate confrontations (bank, widow, bank again) are all built from near-identical rapid one-line question-and-answer volleys, so the chapter reads like three transcripts stapled together rather than three different scenes.
Best The confrontation where Lady Verinder shuts down all talk of her brother is clean, controlled, and quietly tense, letting her authority show through short lines rather than explanation.
Worst It never connects Franklin's early arrival or the sealed packet back to the Indian fortune-tellers and the trance prophecy from earlier that day, so the new mystery feels bolted on rather than grown out of what just happened.
Best It moves briskly through the required plot points without wasted motion or padding.
Worst It reads like a generic mystery scene rather than Christie's book — Poirot loses his usual fussy, theatrical charm and just talks like a plain detective.
Best The very last telegram lands as a genuine, well-timed jolt right after the worrying hospital letter.
Worst The hospital letter has Jonathan already found injured and recovering by 27 May, and a telegram just three days later has him already home in Exeter alive and asking for Mina — this skips straight past the castle imprisonment the story so far says is still going on in June, so it feels like it breaks the story rather than continuing it.
Best It hits every required plot beat cleanly and keeps a consistent late-night, melancholy mood from the Dôme through to the walk home.
Worst Long stretches are built almost entirely from bare one- and two-word lines with no setting or action to anchor them, and the characters say their hidden feelings out loud instead of the story showing them, which flattens both scene and character.
Headline, compliance and craft for each chapter, 0–100.
| Item | Headline | Compliance | Craft | Words |
|---|---|---|---|---|
| Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse | 62.5 | 100 | 62 | 3,070 |
| The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett | 57.4 | 100 | 57 | 2,866 |
| Persuasion item 001 · after ch. 1 · Jane Austen | 56.4 | 100 | 56 | 3,325 |
| The Awakening item 016 · after ch. 7 · Kate Chopin | 50.8 | 95 | 54 | 2,096 |
| The War of the Worlds item 006 · after ch. 1 · H. G. Wells | 50.6 | 95 | 53 | 2,328 |
| Red Harvest item 023 · after ch. 2 · Dashiell Hammett | 49.2 | 95 | 52 | 2,999 |
| The Moonstone item 011 · after ch. 8 · Wilkie Collins | 49.1 | 100 | 49 | 2,538 |
| The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie | 47.9 | 100 | 48 | 2,603 |
| Dracula item 015 · after ch. 5 · Bram Stoker | 42.0 | 85 | 49 | 2,097 |
| The Sun Also Rises item 021 · after ch. 3 · Ernest Hemingway | 36.8 | 100 | 37 | 2,924 |
Words written per item, against each brief’s maximum.
Word count per item vs. brief maximum · run ‘first’
Four of ten items landed inside their brief’s word range. Leave It to Psmith and The Secret Garden and Persuasion and The Awakening and Red Harvest and The Sun Also Rises ran over by 270 and 466 and 725 and 296 and 399 and 324 words respectively.
Each verdict names the chapter’s single worst flaw, with evidence. Quoted verbatim from run ‘first’.
The hospital letter has Jonathan already found injured and recovering by 27 May, and a telegram just three days later has him already home in Exeter alive and asking for Mina — this skips straight past the castle imprisonment the story so far says is still going on in June, so it feels like it breaks the story rather than continuing it.
Long stretches are built almost entirely from bare one- and two-word lines with no setting or action to anchor them, and the characters say their hidden feelings out loud instead of the story showing them, which flattens both scene and character.
Three separate confrontations (bank, widow, bank again) are all built from near-identical rapid one-line question-and-answer volleys, so the chapter reads like three transcripts stapled together rather than three different scenes.
It never connects Franklin's early arrival or the sealed packet back to the Indian fortune-tellers and the trance prophecy from earlier that day, so the new mystery feels bolted on rather than grown out of what just happened.
Automated screening results across the ten outputs.