The Standings → Gemini 3.8 Flash (medium)
medium reasoning effort · via Antigravity CLI · judged by Fable 5 and Gemini 3.7 Flash and GPT-5.6 Sol · the setting shown on the standings; also run at high0.55
Results from run ‘first’: ten items, 30 verdicts.
The record: 10 items · 30 verdicts · 0 errors · 3,596 words on average · est. generation cost $0.57
Six dimensions, each marked 1–5, averaged over ten items and all judges.
Mean craft mark by dimension · 30 verdicts, run ‘first’
Momentum is the strongest dimension (3.60); prose the weakest (3.03). Hover or focus a bar for the per-judge means.
Each item receives one verdict from each judge; the spread of the dots is the disagreement.
Headline score per item, by judge · run ‘first’
| Item | Headline | Fable 5 | Gemini 3.7 Flash | GPT-5.6 Sol | Words |
|---|---|---|---|---|---|
| Dracula item 015 · after ch. 5 · Bram Stoker | 0.68 | 0.71 | 0.79 | 0.55 | 2,779 |
| Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse | 0.63 | 0.59 | 0.75 | 0.54 | 4,256 |
| The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie | 0.59 | 0.59 | 0.71 | 0.47 | 4,246 |
| The War of the Worlds item 006 · after ch. 1 · H. G. Wells | 0.58 | 0.57 | 0.68 | 0.49 | 3,143 |
| The Awakening item 016 · after ch. 7 · Kate Chopin | 0.58 | 0.55 | 0.75 | 0.42† | 2,170 |
| The Sun Also Rises item 021 · after ch. 3 · Ernest Hemingway | 0.56 | 0.62 | 0.62 | 0.42 | 3,514 |
| The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett | 0.55 | 0.60 | 0.68 | 0.37 | 3,512 |
| The Moonstone item 011 · after ch. 8 · Wilkie Collins | 0.53 | 0.50 | 0.60 | 0.49 | 4,887 |
| Red Harvest item 023 · after ch. 2 · Dashiell Hammett | 0.48 | 0.49 | 0.62 | 0.32 | 3,652 |
| Persuasion item 001 · after ch. 1 · Jane Austen | 0.43 | 0.46 | 0.47 | 0.37 | 3,798 |
† The widest judge split in the run: Gemini 3.7 Flash 0.75 against GPT-5.6 Sol 0.42 on The Awakening (item 016).
Words written per item, against each brief’s maximum.
Word count per item vs. brief maximum · run ‘first’
Zero of ten items landed inside their brief’s word range. Dracula and Leave It to Psmith and The Mysterious Affair at Styles and The War of the Worlds and The Awakening and The Sun Also Rises and The Secret Garden and The Moonstone and Red Harvest and Persuasion ran over by 179 and 1,456 and 1,446 and 543 and 370 and 914 and 1,112 and 2,087 and 1,052 and 1,198 words respectively.
Each verdict names the chapter’s single worst flaw, with evidence. Quoted verbatim from run ‘first’.
The trip back for the envelope is visibly engineered to satisfy the brief — Jake insists "I'm not worried about it," then abandons Brett alone at the Select to fetch fifty francs anyway, with Brett herself prompting the errand ("Go back and get your fifty francs"), so the chapter's central movement feels contrived rather than inevitable.
Alfred Inglethorp is played as a transparent melodrama villain — "unctuous," "sanctimoniously," turning "mottled, dirty grey" and stammering with "something like terror" the moment Dorcas contradicts him — which telegraphs guilt so heavily it collapses the misdirection the mystery depends on.
Renfield is made too knowing — his precise \"Miss Lucy Westenra of Chatham Street\" and the prophecy that \"so fair a flower should wither before the summer is through\" over-signal the coming plot and strain the ban on hinting at the supernatural, despite Seward's mundane rationalisation.
The closing telephone call is a transparent exposition device — Mike improbably volunteers exactly the details Psmith needs (the name, the Blandings post, even the bird-of-paradise hat), so the required end state arrives by authorial contrivance rather than by the story's own motion.
Automated screening results across the ten outputs.