Vol. I · No. 1 · MMXXVI
NextChapterBench A Benchmark of Literary Continuation
23 models · 10 items · 1 judge

The Standings → Gemini 3.7 Flash (medium)

Live run data

Google

Gemini 3.7 Flash (medium)

medium reasoning effort · via Antigravity CLI · judged by Claude Sonnet 5 · not the setting shown on the standings; also run at high54.8

Results from run ‘first’: ten items, ten verdicts.

Gemini 3.7 Flash (medium) 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 building scenes, relatively weak on restraint and folding the brief into the story. On the mechanical checks it kept to the brief’s word range on 1 of 10 items and leans on stock phrases more than most.

Headline 48.8 of 100 · compliance × craft
Compliance 92 of 100
Craft 53 of 100

The record: 10 items · 10 verdicts · 0 errors · 3,216 words on average · est. generation cost $0.36

Read the chapters in full

§ I

Craft by Dimension

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

Voice
55
Prose
54
Character
53
Scene
57
Integration
51
Momentum
55
Restraint
41
Specificity
60
0255075100
reader’s score, 0–100 · hollow tick: field average · filled tick: the authors themselves

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.

In plain words, item by item

The judge’s one-sentence best quality and worst flaw for each chapter, written for a general reader.


§ II

Per-Item Results

Headline, compliance and craft for each chapter, 0–100.

The same ten items in figures, with word counts.
Item Headline Compliance Craft Words
Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse 60.390673,619
The Sun Also Rises item 021 · after ch. 3 · Ernest Hemingway 59.8100602,760
Dracula item 015 · after ch. 5 · Bram Stoker 49.595522,440
The Awakening item 016 · after ch. 7 · Kate Chopin 47.795502,104
The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett 47.790533,650
Persuasion item 001 · after ch. 1 · Jane Austen 47.6100483,713
The Moonstone item 011 · after ch. 8 · Wilkie Collins 45.990513,844
The War of the Worlds item 006 · after ch. 1 · H. G. Wells 43.990493,317
The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie 43.390483,357
Red Harvest item 023 · after ch. 2 · Dashiell Hammett 42.375563,351

§ III

Length Against the Brief

Words written per item, against each brief’s maximum.

Word count per item vs. brief maximum · run ‘first’

words written brief maximum
Leave It to Psmith
3,619
The Sun Also Rises
2,760
Dracula
2,440
The Awakening
2,104
The Secret Garden
3,650
Persuasion
3,713
The Moonstone
3,844
The War of the Worlds
3,317
The Mysterious Affair at Styles
3,357
Red Harvest
3,351
01,0002,0003,0004,000
words; the ink tick marks each brief’s maximum

One of ten items landed inside their brief’s word range. Leave It to Psmith and The Sun Also Rises and The Awakening and The Secret Garden and Persuasion and The Moonstone and The War of the Worlds and The Mysterious Affair at Styles and Red Harvest ran over by 819 and 160 and 304 and 1,250 and 1,113 and 1,044 and 717 and 557 and 751 words respectively.


§ IV

Worst Flaws, as Recorded

Each verdict names the chapter’s single worst flaw, with evidence. Quoted verbatim from run ‘first’.

It breaks the single most important rule in the brief by bringing Miss Rachel onstage for a whole dinner scene, laughing and dressed up, when she was supposed to stay offpage, and it oversells danger with a melodramatic gunpowder comparison and a speech from Lady Verinder that explains her own feelings out loud instead of showing them.

on The Moonstone, item 011 · Claude Sonnet 5

The prose keeps over-explaining feelings and clues with purple flourishes ('his eyes shone with that strange, emerald light,' 'the good God has given us our eyes not merely to look with, but to see') that tell the reader what to feel instead of trusting the scene.

on The Mysterious Affair at Styles, item 017 · Claude Sonnet 5

It gives the anonymous narrator a fixed name, 'Neill,' used openly with strangers, and lets the widow's fear spill out in overwrought physical description (gasping, whitened knuckles, dropped handbag), which breaks the cool, closed-off voice of the original.

on Red Harvest, item 023 · Claude Sonnet 5

Psmith actually tells Walderwick the whole story of the umbrella instead of dodging the question, and the ending where he tucks her note next to his heart feels sentimental in a way this dry, joking narrator wouldn't normally allow.

on Leave It to Psmith, item 022 · Claude Sonnet 5

§ V

The Mechanical Docket

Automated screening results across the ten outputs.