The Standings → Gemini 3.8 Flash
high reasoning effort · via Antigravity CLI · judged by Claude Sonnet 5 · not the setting shown on the standings; also run at medium54.6
Results from run ‘first’: ten items, ten verdicts.
Gemini 3.8 Flash scores 53 on craft, 10 points below the field’s 63 and 40 points short of the authors themselves. Its shape differs from the field’s: relatively strong on period detail and momentum, relatively weak on restraint and folding the brief into the story. On the mechanical checks it kept to the brief’s word range on 5 of 10 items and leans on stock phrases more than most.
The record: 10 items · 10 verdicts · 0 errors · 3,092 words on average · est. generation cost $1.67
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 53. 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 Renfield interview is genuinely well-built, piling up concrete detail (the newspaper fly-boxes, the sly rubbing of hands) before the uncanny question lands.
Worst It leans on stock horror phrases like an 'icy hand' gripping the heart and a sudden 'brain fever' twist that tell us how to feel instead of letting the scene do the work.
Best The confrontation has a clean, easy-to-follow shape with real numbers attached to each proposal, so the stakes of the money trouble feel concrete.
Worst Every strong emotion is spelled out with stock physical description (faces flushing crimson, eyes flashing with fury, voices thundering) instead of the dry, understated irony the source chapter uses.
Best Dorcas's flat, dignified contradiction of Alfred's alibi ('I know what I did with my own hands, sir') is a genuinely well-staged moment of quiet servant courage against a blustering gentleman.
Worst The prose keeps stopping to editorialize about how terrible and strange everything is ('How strange it seemed that nature could be so calm...'), which drains tension instead of building it and doesn't sound like the clipped, restrained way this narrator actually talks.
Best It hits every required plot point cleanly and gives a nice concrete detail (Léonce's telegram coming from Carondelet Street) that fits the world of the story.
Worst The quiet, praying lady in black suddenly turns into an ominous fortune-teller who spells out the story's watery symbolism in plain words, which feels too obvious for how she's been written up to now.
Best The bank scene is tight and procedural, and small touches like Thaler's note needling the Op about his weight show real care in tracking details from the source chapter.
Worst It tips into melodrama the source never uses — the widow flatly threatens 'you are a dead man' and Thaler's note comes with a cartoonish drawn symbol, which feels more like a pulp thriller than the flat, dry menace of the original.
Best The plotting is brisk and clear, with a genuinely creepy anonymous letter and a well-staged locking-away of the packet that keeps the reader oriented in the house's geography.
Worst The writing keeps tipping into stock thriller melodrama — grabbing collars, 'walking in the shadow of death,' 'the devils are here already!' — which reads more like a Victorian penny-dreadful pastiche than Collins' dry, self-mocking narrator.
Best It hits every required story beat cleanly and keeps the action moving at a brisk, readable pace.
Worst It piles on melodramatic description (Craven's 'brow deeply scored with bitter grief', the wind that 'pounced upon her like an immense creature', Craven driving off 'like a phantom') that feels more like an adventure novel than Burnett's plain, withholding style.
Best The scene where Psmith calmly buys the stolen umbrella back off its furious owner for twenty pounds is a well-built, escalating comic negotiation.
Worst It crams in too many new plot gadgets at once (an ink spill, a written questionnaire, a phone call that recaps the whole day) so the comedy feels busy rather than sharp, and it slips by having a friend call Psmith 'Rupert,' which is actually the villain-secretary Baxter's name.
Best It hits every required story beat cleanly and lands the ending almost exactly on the required moment, with the noon bell striking as the wife disappears and the crowd swells.
Worst It piles on melodrama and over-explained fear (exclamation points, the wife calling the cylinder 'evil, like an engine of war,' the narrator sensing 'great peril'), which breaks the calm, understated tone the source establishes.
Best The scene where Jake finds Georgette still waiting at the bal is sharp, with believable, weary dialogue for her.
Worst The writing keeps explaining exactly how Jake feels instead of trusting the reader, and it piles on flowery description that doesn't sound like this author's plain style.
Headline, compliance and craft for each chapter, 0–100.
| Item | Headline | Compliance | Craft | Words |
|---|---|---|---|---|
| Dracula item 015 · after ch. 5 · Bram Stoker | 57.4 | 100 | 57 | 2,508 |
| Persuasion item 001 · after ch. 1 · Jane Austen | 55.0 | 100 | 55 | 2,457 |
| The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie | 54.0 | 95 | 57 | 4,580 |
| The Awakening item 016 · after ch. 7 · Kate Chopin | 53.5 | 100 | 54 | 1,730 |
| Red Harvest item 023 · after ch. 2 · Dashiell Hammett | 51.9 | 95 | 55 | 2,860 |
| The Moonstone item 011 · after ch. 8 · Wilkie Collins | 51.9 | 90 | 58 | 3,048 |
| The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett | 51.4 | 100 | 51 | 2,311 |
| Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse | 49.9 | 95 | 52 | 4,717 |
| The War of the Worlds item 006 · after ch. 1 · H. G. Wells | 40.9 | 85 | 48 | 2,559 |
| The Sun Also Rises item 021 · after ch. 3 · Ernest Hemingway | 40.0 | 95 | 42 | 4,146 |
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. The Mysterious Affair at Styles and Red Harvest and The Moonstone and Leave It to Psmith and The Sun Also Rises ran over by 1,780 and 260 and 248 and 1,917 and 1,546 words respectively.
Each verdict names the chapter’s single worst flaw, with evidence. Quoted verbatim from run ‘first’.
It crams in too many new plot gadgets at once (an ink spill, a written questionnaire, a phone call that recaps the whole day) so the comedy feels busy rather than sharp, and it slips by having a friend call Psmith 'Rupert,' which is actually the villain-secretary Baxter's name.
The prose keeps stopping to editorialize about how terrible and strange everything is ('How strange it seemed that nature could be so calm...'), which drains tension instead of building it and doesn't sound like the clipped, restrained way this narrator actually talks.
It piles on melodramatic description (Craven's 'brow deeply scored with bitter grief', the wind that 'pounced upon her like an immense creature', Craven driving off 'like a phantom') that feels more like an adventure novel than Burnett's plain, withholding style.
The writing keeps tipping into stock thriller melodrama — grabbing collars, 'walking in the shadow of death,' 'the devils are here already!' — which reads more like a Victorian penny-dreadful pastiche than Collins' dry, self-mocking narrator.
Automated screening results across the ten outputs.