NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-17-01KHMMC989KMWXK522CDGRHB1G
Run Details
Model:
Mistral Large
Test:
Bad Writing Habits
Scenario:
Thriller: chase through city streets
Duration:
26.9s
Cost:
$0.009852
Model input
System
You are an expert novelist with a gift for vivid, immersive prose. You write in tight third-person limited point of view with a strong narrative voice. Your prose is clean, precise, and engaging — you favor strong verbs, concrete details, and natural-sounding dialogue. You will be given a story bible (codex) containing canonical details about characters, locations, items, and lore. Use these details to ground your writing in the established world. Stay consistent with the codex. Write only the prose of the scene. Do not include titles, chapter headings, markdown formatting, or any commentary before or after the scene.
User/Human
<codex> <character id="4416" name="Detective Harlow Quinn" aliases="Quinn, Detective Quinn, Harlow" pronouns="she/her" species="Human" storyRole="Antagonist"> <eyeColor>brown</eyeColor> <hair>closely cropped salt-and-pepper</hair> <height>5'9"</height> <age>41</age> <jawline>sharp jaw</jawline> <watch>worn leather watch on her left wrist</watch> <bearing>military precision</bearing> <description>A Metropolitan Police detective with 18 years of decorated service who suspects the clique is involved in criminal activity.</description> <backstory>She lost her partner, DS Morris, during a case three years ago under unexplained circumstances that had supernatural origins she doesn't yet understand.</backstory> </character> <character id="4417" name="Tomás Herrera" aliases="Tommy, Tomás, Herrera" pronouns="he/him" species="Human" storyRole="Support"> <eyeColor>warm brown</eyeColor> <hair>short curly dark brown</hair> <height>5'10"</height> <age>29</age> <skinTone>olive skin</skinTone> <scar>scar running along his left forearm from a knife attack</scar> <medallion>Saint Christopher medallion around his neck</medallion> <birthplace>Seville, Spain</birthplace> <description>A former paramedic who provides off-the-books medical care for the clique.</description> <backstory>Born in Seville, Spain. He moved to London to work for the NHS but lost his license after administering unauthorized treatments to supernatural patients.</backstory> </character> <location id="854" name="The Raven's Nest" aliases="the Nest, Raven, Silas' bar" realm="Earth"> <signColor>distinctive green neon sign above the entrance</signColor> <secretRoom>hidden back room accessible through a bookshelf, used for clandestine meetings</secretRoom> <description>A dimly lit bar in Soho, London. The walls are covered with old maps and black-and-white photographs.</description> </location> <location id="856" name="The Veil Market" aliases="Veil Market, the Market, underground market" realm="Earth"> <currentLocation>abandoned Tube station beneath Camden</currentLocation> <entryRequirement>bone token</entryRequirement> <schedule>moves locations every full moon</schedule> <description>A hidden supernatural black market that sells enchanted goods, banned alchemical substances, and information.</description> </location> </codex> Write a fully fleshed out scene of at least 1000 words in the thriller genre. A detective is pursuing a suspect through rain-soaked city streets at night. The chase takes them from the street level down into an underground market. The detective must decide whether to follow the suspect into unfamiliar and potentially dangerous territory. Use the codex entries above as your world reference for characters and setting. Only include the prose of the scene without any markdown or commentary.
Model response
Messages
Meta
AI
The rain came down in sheets, turning the London streets into a slick, reflective maze. Detective Harlow Quinn moved with military precision, her boots splashing through puddles as she kept her eyes locked on the figure ahead. Tomás Herrera—*Tommy*—was fast, his dark curls plastered to his forehead, his olive skin
glistening
under the
flicker
ing streetlights. He glanced back, just once, and Quinn saw the panic in his warm brown eyes before he ducked into a narrow alley. She didn’t hesitate. The alley reeked of wet pavement and stale beer, the walls slick with grime. Quinn’s breath came sharp and controlled, her fingers brushing the worn leather strap of her
watch
as she adjusted her grip on the slick railing of a fire escape. Herrera was already halfway up, his Saint Christopher medallion bouncing against his chest with each desperate climb. She followed, her sharp jaw set, salt-and-pepper hair damp against her scalp. At the top, Herrera vaulted over a low wall, landing hard on the other side. Quinn didn’t break stride. She cleared the wall with practiced ease, her boots hitting the ground with a wet *thud*. The rooftops stretched ahead, a jagged skyline of chimneys and satellite dishes, all
glistening
under the storm. Herrera
was heading
for the edge—toward the next building, toward the drop. Quinn’s pulse hammered in her throat. She’d chased enough suspects to know when one
was cornered
. And Herrera
was running
like a man who had nowhere left to go. Then he disappeared. Not over the edge. Not down a drainpipe. He was just… *gone*. Quinn skidded to a stop at the rooftop’s lip, her breath fogging in the cold air. Below, the street was a blur of headlights and neon, the rain distorting the shapes into something almost
surreal
. But there, at the mouth of an alley, was a
flicker
of movement—a shadow where there shouldn’t have been one. She descended the fire escape two steps at a time, her mind racing. Herrera wasn’t just a suspect. He was connected—*too* connected—to the clique, to the cases that had been eating at her for months. The ones that didn’t add up. The ones that left bodies drained of blood, or worse, *changed*. The alley ended at a rusted metal door, half-hidden behind a tangle of overgrown ivy. It was ajar, the hinges groaning
softly
in the wind. Quinn hesitated for the first time. The air here was different—thicker, charged, like the moment before a storm broke. She could turn back. Call for backup. Follow protocol. But protocol had never brought her closer to the truth about what
happened to
Morris. She pushed the door open. The stairs beyond were narrow and steep, the walls damp and covered in peeling paint. The air smelled of earth and something older—something metallic and faintly sweet, like copper and ozone. Quinn descended
slowly
, her hand resting on the grip of her service weapon. The steps spiraled downward, deeper than any basement should have gone, until they opened into a cavernous space. The Veil Market. Quinn had heard whispers of it—rumors of a black market that moved with the moon, selling things that shouldn’t exist. But seeing it was different. The abandoned Tube station had
been transformed
. Stalls lined the tracks, their wares glowing under the dim,
flicker
ing lights: vials of shimmering liquid, bundles of herbs that pulsed with an eerie light, weapons that hummed with something like electricity. The air thrummed with voices, low and urgent, the language a mix of tongues she didn’t recognize. And there, weaving through the crowd, was Herrera. Quinn stepped forward, her boots silent on the damp stone. The market’s patrons barely glanced at her—some were human, or close enough, but others… others had eyes that reflected the light wrong, or skin that shifted like smoke. She kept her gaze locked on Herrera, her fingers flexing around the grip of her gun. He
was heading
for a stall at the far end, one draped in black
silk
. The vendor, a gaunt figure with too many fingers, handed him something small and pale—a bone token, Quinn realized with a jolt. Herrera clutched it in his fist, then turned toward a shadowed archway at the back of the station. Quinn followed. The archway led to a tunnel, the walls lined with
flicker
ing torches that cast long,
wavering
shadows. The air grew colder with each step, her breath misting in front of her. Herrera was ahead, his pace slowing, his shoulders tense. He knew she was there. “Herrera!” Quinn’s voice
echoed
down the tunnel. “Metropolitan Police. Stop where I can see you.” He didn’t stop. But he did slow, his hands raising
slightly
, the bone token still clutched in his fingers. “You shouldn’t have followed me, Detective.” “Why’s
that?” Quinn
kept her voice steady, her eyes scanning the tunnel. The walls here were different—smooth, almost polished, like the inside of a vein. “What’s down here, Herrera?” He turned then, his warm brown eyes wide, his scarred forearm tense. “You don’t want to know.” “Try me.” Herrera exhaled
sharply
, his breath unsteady. “You’re already in too deep. You just don’t know it yet.” He held up the token. “This is the only way out. But you can’t take it from me. You have to *earn* it.” Quinn’s grip tightened on her gun. “What the hell does that mean?” Before Herrera could answer, the torches
flickered
. The shadows at the edges of the tunnel *moved*, stretching toward them like living things. A low, guttural sound
echoed
from the darkness, something between a growl and a whisper. Herrera’s face paled. “We’re out of time.” Quinn didn’t wait for an explanation. She lunged forward, grabbing for the token. Herrera twisted away, but not fast enough—her fingers closed around his wrist, the bone cold against her skin. The shadows surged forward, and for a heartbeat, Quinn saw *them*—pale, gaunt faces in the dark, eyes black and endless. Then Herrera yanked her forward, into the unknown.
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
0
adverbTagCount
0
adverbTags
(empty)
60.2386%
AI-ism adverb frequency
Target: <1% AI-ism adverbs (57 tracked)
wordCount
1006
totalAiIsmAdverbs
4
found
0
adverb
"sharply"
count
1
1
adverb
"slightly"
count
1
2
adverb
"slowly"
count
1
3
adverb
"softly"
count
1
highlights
0
"sharply"
1
"slightly"
2
"slowly"
3
"softly"
100.0000%
AI-ism character names
Target: 0 AI-default names (17 tracked, −20% each)
codexExemptions
(empty)
found
(empty)
100.0000%
AI-ism location names
Target: 0 AI-default location names (33 tracked, −20% each)
codexExemptions
(empty)
found
(empty)
10.5368%
AI-ism word frequency
Target: <1% AI-ism words (140 tracked)
wordCount
1006
totalAiIsms
9
found
0
word
"flickered"
count
1
1
word
"flicker"
count
1
2
word
"glistening"
count
2
3
word
"wavering"
count
1
4
word
"surreal"
count
1
5
word
"silk"
count
1
6
word
"echoed"
count
2
highlights
0
"flickered"
1
"flicker"
2
"glistening"
3
"wavering"
4
"surreal"
5
"silk"
6
"echoed"
100.0000%
Cliche density
Target: ≤1 cliche(s) per 800-word window
totalCliches
0
maxInWindow
0
found
(empty)
highlights
(empty)
100.0000%
Emotion telling (show vs. tell)
Target: ≤3% sentences with emotion telling
emotionTells
0
narrationSentences
80
matches
(empty)
100.0000%
Filter word density
Target: ≤12% sentences with filter/hedge words
filterCount
1
hedgeCount
1
narrationSentences
80
filterMatches
0
"watch"
hedgeMatches
0
"happened to"
100.0000%
Overuse of "that" (subordinate clause padding)
Target: ≤10% sentences with "that" clauses
thatCount
1
totalSentences
88
matches
0
"that?” Quinn"
100.0000%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
31
mean
32.03
std
25.05
cv
0.782
sampleLengths
0
77
1
3
2
72
3
64
4
29
5
3
6
12
7
55
8
52
9
53
10
15
11
5
12
62
13
3
14
81
15
8
16
54
17
55
18
2
19
45
20
15
21
25
22
29
23
17
24
2
25
40
26
12
27
37
28
7
29
51
30
8
100.0000%
Passive voice overuse
Target: ≤5% passive sentences
passiveCount
2
totalSentences
80
matches
0
"was cornered"
1
"been transformed"
100.0000%
Past progressive (was/were + -ing) overuse
Target: ≤10% past progressive verbs
pastProgressiveCount
3
totalVerbs
156
matches
0
"was heading"
1
"was running"
2
"was heading"
94.0000%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
939
adjectiveStacks
1
stackExamples
0
"different—thicker, charged, like"
adverbCount
33
adverbRatio
0.03514376996805112
lyAdverbCount
6
lyAdverbRatio
0.006389776357827476
100.0000%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
88
echoCount
0
echoWords
(empty)
100.0000%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
88
mean
11.28
std
6.53
cv
0.579
sampleLengths
0
15
1
22
2
18
3
22
4
3
5
14
6
29
7
17
8
12
9
15
10
4
11
16
12
17
13
12
14
6
15
10
16
13
17
3
18
4
19
4
20
3
21
1
22
16
23
19
24
20
25
13
26
5
27
17
28
6
29
11
30
15
31
10
32
6
33
13
34
4
35
3
36
2
37
15
38
5
39
15
40
16
41
13
42
18
43
3
44
20
45
5
46
7
47
32
48
17
49
8
46.2121%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
6
diversityRatio
0.3181818181818182
totalSentences
88
uniqueOpeners
28
100.0000%
Dialogue tag variety (said vs. fancy)
Target: ≤30% fancy dialogue tags
totalTags
0
fancyCount
0
fancyTags
(empty)
88.8816%