NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-02-17-01KHMR15JVDNHRFCN53M09S1A0
Run Details
Model:
Gemini 2.5 Pro
Test:
Bad Writing Habits
Scenario:
Thriller: chase through city streets
Duration:
38.6s
Cost:
$0.037845
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
100.00%
Adverbs in dialogue tags
(details)
93.39%
AI-ism adverb frequency
(details)
100.00%
AI-ism character names
(details)
100.00%
AI-ism location names
(details)
30.56%
AI-ism word frequency
(details)
100.00%
Cliché density
(details)
100.00%
Emotion telling (show vs. tell)
(details)
100.00%
Filter word density
(details)
100.00%
Gibberish response detection
(details)
100.00%
Markdown formatting overuse
(details)
100.00%
Missing dialogue indicators (quotation marks)
(details)
100.00%
Name drop frequency
(details)
50.00%
Narrator intent-glossing
(details)
100.00%
"Not X but Y" pattern overuse
(details)
100.00%
Overuse of "that" (subordinate clause padding)
(details)
100.00%
Paragraph length variance
(details)
96.84%
Passive voice overuse
(details)
7.96%
Past progressive (was/were + -ing) overuse
(details)
100.00%
Em-dash & semicolon overuse
(details)
94.00%
Purple prose (modifier overload)
(details)
100.00%
Repeated phrase echo
(details)
100.00%
Sentence length variance
(details)
33.20%
Sentence opener variety
(details)
0.00%
Adverb-first sentence starts
(details)
48.57%
Pronoun-first sentence starts
(details)
35.63%
Subject-first sentence starts
(details)
42.02%
Subordinate conjunction sentence starts
(details)
100.00%
Technical jargon density
(details)
100.00%
Useless dialogue additions
(details)
100.00%
Dialogue tag variety (said vs. fancy)
(details)
81.0719%
Overall
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
Rain sluiced down the windscreen, turning the Soho night into a watercolour of bleeding neon.
Detective Harlow Quinn killed the
engine, the wipers sighing to a halt.
The quiet cocoon of the
car amplified the drumming on the roof. Across the street, the distinctive green sign of The Raven’s Nest sizzled, its reflection writhing on the wet asphalt. For three hours, she had
watched that sign
, a fixed point in the storm, letting the
familiar
ache settle into her bones.
Patience was a muscle, and
after eighteen years on the force, hers
was honed
to a razor’s edge.
A figure pushed through the
bar’s heavy door, hunching his shoulders against the downpour. Tomás Herrera.
Quinn’s focus narrowed, the rest
of the street dissolving. He matched the file
perfectly
: short, curly dark hair already plastered to his skull, olive skin pale under the lurid green light.
He clutched the collar of
his jacket, his warm brown eyes
scanning
the street with a frantic energy that had nothing to do with the weather.
He wasn’t looking for a
cab.
He was looking for a
shadow. And Quinn was it.
Herrera set off, his pace
a clipped, urgent walk that threatened to break into a run.
Quinn gave him a ten-second
lead before easing her door open.
The cold shock of the
rain hit her instantly, plastering the cropped salt-and-pepper strands to her scalp.
She pulled up the collar
of her own coat, the
familiar
scent of wet wool and London air filling her senses.
Her worn leather watch felt
tight on her wrist, its steady tick a counterpoint to her
pounding
heart.
She moved with the military
precision that had never left her, a ghost in the deluge.
Her footsteps were silent splashes
on the pavement, her gaze locked on Herrera’s back sixty feet ahead.
He navigated the crowded pavement
with jerky, desperate movements, a
stark
contrast to her own fluid economy of motion.
He was prey, and he
knew it.
He risked a glance over
his shoulder, and for a half-second, their eyes met across the rain-swept street. Panic flared in his face, bright and raw. He ran. Quinn didn’t hesitate.
She lengthened her stride, the
ache in her knees a dull
throb
. The chase was on. Concrete and glass blurred past.
She vaulted a low barrier
separating the pavement from the road, a car horn blaring in protest as she cut across the street. Herrera was younger, but he ran on fear. Quinn ran on purpose.
She saw the route in
her head, a strategic map of alleys and shortcuts.
He ducked into a narrow
passage between a theatre and a bookshop, and she followed without breaking her stride. The noise of the city muted, replaced by the
echo
of their footfalls and the gurgle of water rushing through a drainpipe. The air grew thick with the smell of wet brick and refuse. Herrera scrambled over a stack of collapsed cardboard boxes, sending them sliding across the slick cobblestones. Quinn sidestepped the mess, her eyes never leaving him.
His breath came in ragged
bursts she could hear even over the storm.
He fumbled with the clasp
on a metal gate, his hands shaking. The gate screeched open.
He slipped through, trying to
slam it shut behind him, but Quinn was too close.
She slammed her shoulder into
the latticed steel, the impact jarring through her frame. The gate flew back, striking the wall with a deafening clang. Herrera stumbled, catching himself on the brickwork.
He spared her another terrified
look before bolting again, back out onto a wider, emptier street heading north, towards Camden. He
was leading
her away from the crowded heart of the city, towards the quieter, more derelict fringes. A tactic. Or a destination. Quinn pushed the thought aside, focusing on the burn in her lungs, the rhythm of the pursuit. She
was gaining
. The gap between them closed from twenty feet to ten. She could see the details now: the faded Saint Christopher medallion bouncing against his chest, the jagged white line of a scar on his left forearm. He was a medic, not a professional runner. He
was running
out of steam. He veered
suddenly
, making for a graffiti-scrawled service entrance set into the tiled wall of a Tube underpass. It was an old, forgotten door, streaked with grime and rust. He reached it, his back pressed against the metal as he fumbled in his pocket. He pulled out a small, pale object. In the low light, it
looked like a piece of carved bone
. He pressed it against a spot on the door Quinn couldn't see. There was no click, no sound of a lock disengaging, but the door swung silently inward. He threw himself through the opening, and the door swung shut with a heavy, final thud. Quinn reached it two seconds later. She grabbed the handle, a solid, unmoving bar of cold iron. Locked. Cursing under her breath, she pulled a small
tension
wrench and a pick from an inner pocket. Her fingers, numb with cold, felt clumsy, but the
familiar
motions steadied her. This was her world. Locks, doors, and the secrets they kept. She worked by feel, her ear pressed to the cold metal, listening for the faint clicks of the tumblers. The lock was old but
complex
. Not
standard
issue. It gave way with a soft snap that was more felt than heard. She pushed the door open and stepped inside, gun drawn. The world changed. The roar of the rain vanished, swallowed by a
profound
, humming
silence
. The air tasted of ozone, damp earth, and something else… a faint, cloying sweetness like overripe fruit. She was in a disused Tube tunnel. A string of emergency lights cast a sickly yellow glow down a long, curved platform, illuminating rails that disappeared into an abyss of black. Water dripped from the arched ceiling, each drop
echoing
in the vast space. Herrera
was gone
. A set of wet footprints led away from the door, down the platform, and into the darkness. Quinn followed, her
footsteps
loud against the concrete. The only other sound was the steady drip, drip, drip from above and the low, resonant
hum that seemed to vibrate up through the soles of her shoes
. She
felt a prickle
of unease, a cold dread that had nothing to do with the damp air. It was a feeling she hadn't experienced in three years. Not since the night she’d found DS Morris in that warehouse, the air thick with the same cloying sweetness, his eyes wide with a terror she could never decipher. She rounded a wide curve in the tunnel, and the humming grew louder. Light spilled from an archway ahead,
not the sterile yellow of the emergency lamps, but a shifting, vibrant kaleidoscope of colours
. And with the light came noise. A low murmur of voices, strange music, the chittering of creatures she couldn't name. Quinn slowed, pressing herself against the damp, curved wall. She peered around the edge of the archway. It wasn't a tunnel. It was a cavernous station, a long-forgotten stop on a dead line, and it was alive. The platform
was packed
with stalls, lit by glowing crystals and flitting balls of light. A market. A crowd of beings milled between the vendors, their forms and faces a dizzying, impossible
mosaic
. A man with skin like polished obsidian bargained with a vendor whose hands ended in delicate, feathery talons. A woman in a
silk
robe laughed, her teeth a row of sharpened pearls. Cages held creatures that skittered and glowed, and jars on shelves contained swirling nebulae of coloured dust. This was where Herrera had run. This was his world. Her gaze swept the impossible scene, and she saw him. Across the bustling platform, near a stall selling what looked like bottled lightning, Tomás Herrera
was talking
animatedly to a tall figure cloaked in shadow. He gestured back the way he had come, his expression still laced with fear. He
was seeking
refuge. Or reporting in. Quinn’s
blood ran cold
. Every instinct screamed at her to retreat, to call for backup that would never believe her, to seal the door and pretend this place didn't exist. This was the unexplained. This was the world that had swallowed her partner whole.
To step through that archway
was to leave behind everything she knew, every rule and regulation that gave her work meaning. Her authority was a worthless token here. Her gun
felt like a child’s toy
. She looked at Herrera, at the secrets he so clearly harboured. The answers she'd been chasing for three years felt impossibly close, just on the other side of that threshold. A choice. Turn back to the rain and the
familiar
, broken world, or step into the abyss that took Morris. Her sharp jaw tightened. Her hand was steady as she checked the safety on her weapon. Taking a breath that tasted of ozone and magic, Detective Harlow Quinn stepped out of the shadows and into the light of the Veil Market.
created_at
"2026-02-17T02:50:00.022513+00:00"
2/17/2026, 2:50:00 AM
model
"google/gemini-2.5-pro"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2183
moderation_latency
(null)
generation_time
38555
tokens_prompt
914
tokens_completion
3168
native_tokens_prompt
852
native_tokens_completion
3678
native_tokens_completion_images
0
native_tokens_reasoning
1776
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"STOP"
usage
0.037845
router
(null)
provider_responses
0
status
200
is_byok
false
latency
2183
endpoint_id
"b261d15b-95ce-4c8c-8bce-00eb0dc1ff77"
provider_name
"Google AI Studio"
model_permaslug
"google/gemini-2.5-pro"
api_type
"completions"
id
"gen-1771296560-XXvCXRFattQwMidl5ooS"
upstream_id
(null)
total_cost
0.037845
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
0
93.39%
AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount
1512
totalAiIsmAdverbs
2
found
0
adverb
"perfectly"
count
1
1
adverb
"suddenly"
count
1
highlights
0
"perfectly"
1
"suddenly"
100.00%
AI-ism character names
Target: 0 AI-default names (17 tracked, −20% each)
codexExemptions
(empty)
found
(empty)
100.00%
AI-ism location names
Target: 0 AI-default location names (33 tracked, −20% each)
codexExemptions
(empty)
found
(empty)
30.56%
AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount
1512
totalAiIsms
21
found
0
word
"familiar"
count
4
1
word
"scanning"
count
1
2
word
"pounding"
count
1
3
word
"footsteps"
count
2
4
word
"navigated"
count
1
5
word
"stark"
count
1
6
word
"throb"
count
1
7
word
"echo"
count
1
8
word
"tension"
count
1
9
word
"complex"
count
1
10
word
"standard"
count
1
11
word
"profound"
count
1
12
word
"silence"
count
1
13
word
"echoing"
count
1
14
word
"kaleidoscope"
count
1
15
word
"mosaic"
count
1
16
word
"silk"
count
1
highlights
0
"familiar"
1
"scanning"
2
"pounding"
3
"footsteps"
4
"navigated"
5
"stark"
6
"throb"
7
"echo"
8
"tension"
9
"complex"
10
"standard"
11
"profound"
12
"silence"
13
"echoing"
14
"kaleidoscope"
15
"mosaic"
16
"silk"
100.00%
Cliché density
Target: ≤1 cliche(s) per 800-word window
totalCliches
1
maxInWindow
1
found
0
label
"blood ran cold"
count
1
highlights
0
"blood ran cold"
100.00%
Emotion telling (show vs. tell)
Target: ≤3% sentences with emotion telling
emotionTells
2
narrationSentences
125
matches
0
"felt a prickle"
1
"d with fear"
100.00%
Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount
1
hedgeCount
1
narrationSentences
125
filterMatches
0
"watch"
hedgeMatches
0
"seemed to"
100.00%
Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences
125
gibberishSentences
0
adjustedGibberishSentences
0
longSentenceCount
0
runOnParagraphCount
0
giantParagraphCount
0
wordSaladCount
0
repetitionLoopCount
0
controlTokenCount
0
maxSentenceWordsSeen
29
ratio
0
matches
(empty)
100.00%
Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans
0
markdownWords
0
totalWords
1512
ratio
0
matches
(empty)
100.00%
Missing dialogue indicators (quotation marks)
Target: ≤10% speech attributions without quotation marks
totalAttributions
2
unquotedAttributions
0
matches
(empty)
100.00%
Name drop frequency
Target: ≤1.0 per-name mentions per 100 words
totalMentions
44
wordCount
1512
uniqueNames
16
maxNameDensity
0.99
worstName
"Quinn"
maxWindowNameDensity
2
worstWindowName
"Quinn"
discoveredNames
Soho
1
Harlow
2
Quinn
15
Raven
1
Nest
1
Herrera
10
London
1
Camden
1
Saint
1
Christopher
1
Tube
2
Morris
2
Tomás
2
Detective
2
Veil
1
Market
1
persons
0
"Harlow"
1
"Quinn"
2
"Raven"
3
"Nest"
4
"Herrera"
5
"Saint"
6
"Christopher"
7
"Morris"
8
"Tomás"
places
0
"Soho"
1
"London"
2
"Camden"
globalScore
1
windowScore
1
50.00%
Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences
100
glossingSentenceCount
4
matches
0
"looked like a piece of carved bone"
1
"hum that seemed to vibrate up through the soles of her shoes"
2
"looked like bottled lightning, Tomás Herr"
3
"felt like a child’s toy"
100.00%
"Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches
1
per1kWords
0.661
wordCount
1512
matches
0
"not the sterile yellow of the emergency lamps, but a shifting, vibrant kaleidoscope of colours"
100.00%
Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount
1
totalSentences
125
matches
0
"watched that sign"
100.00%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
22
mean
68.73
std
35.09
cv
0.511
sampleLengths
0
98
1
85
2
4
3
83
4
85
5
2
6
92
7
84
8
71
9
93
10
107
11
100
12
10
13
76
14
113
15
54
16
119
17
10
18
56
19
79
20
50
21
41
96.84%
Passive voice overuse
Target: ≤2% passive sentences
passiveCount
3
totalSentences
125
matches
0
"was honed"
1
"was gone"
2
"was packed"
7.96%
Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount
7
totalVerbs
243
matches
0
"wasn’t looking"
1
"was looking"
2
"was leading"
3
"was gaining"
4
"was running"
5
"was talking"
6
"was seeking"
100.00%
Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount
0
semicolonCount
0
flaggedSentences
0
totalSentences
125
ratio
0
matches
(empty)
94.00%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
1519
adjectiveStacks
1
stackExamples
0
"short, curly dark hair"
adverbCount
26
adverbRatio
0.017116524028966424
lyAdverbCount
10
lyAdverbRatio
0.0065832784726793945
100.00%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
125
echoCount
0
echoWords
(empty)
100.00%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
125
mean
12.1
std
6.77
cv
0.56
sampleLengths
0
15
1
12
2
12
3
19
4
22
5
18
6
14
7
2
8
9
9
22
10
26
11
6
12
6
13
4
14
16
15
11
16
17
17
21
18
18
19
16
20
17
21
19
22
7
23
18
24
8
25
2
26
3
27
12
28
4
29
5
30
23
31
8
32
4
33
14
34
19
35
22
36
12
37
16
38
9
39
13
40
12
41
4
42
15
43
14
44
11
45
7
46
20
47
18
48
2
49
3
33.20%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
21
diversityRatio
0.264
totalSentences
125
uniqueOpeners
33
0.00%
Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount
0
totalSentences
119
matches
(empty)
ratio
0
48.57%
Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount
51
totalSentences
119
matches
0
"He matched the file perfectly:"
1
"He clutched the collar of"
2
"He wasn’t looking for a"
3
"He was looking for a"
4
"She pulled up the collar"
5
"Her worn leather watch felt"
6
"She moved with the military"
7
"Her footsteps were silent splashes"
8
"He navigated the crowded pavement"
9
"He was prey, and he"
10
"He risked a glance over"
11
"She lengthened her stride, the"
12
"She vaulted a low barrier"
13
"She saw the route in"
14
"He ducked into a narrow"
15
"His breath came in ragged"
16
"He fumbled with the clasp"
17
"He slipped through, trying to"
18
"She slammed her shoulder into"
19
"He spared her another terrified"
ratio
0.429
35.63%
Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount
101
totalSentences
119
matches
0
"Rain sluiced down the windscreen,"
1
"Detective Harlow Quinn killed the"
2
"The quiet cocoon of the"
3
"Patience was a muscle, and"
4
"A figure pushed through the"
5
"Quinn’s focus narrowed, the rest"
6
"He matched the file perfectly:"
7
"He clutched the collar of"
8
"He wasn’t looking for a"
9
"He was looking for a"
10
"Herrera set off, his pace"
11
"Quinn gave him a ten-second"
12
"The cold shock of the"
13
"She pulled up the collar"
14
"Her worn leather watch felt"
15
"She moved with the military"
16
"Her footsteps were silent splashes"
17
"He navigated the crowded pavement"
18
"He was prey, and he"
19
"He risked a glance over"
ratio
0.849
42.02%
Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount
1
totalSentences
119
matches
0
"To step through that archway"
ratio
0.008
100.00%
Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences
77
technicalSentenceCount
4
matches
0
"He clutched the collar of his jacket, his warm brown eyes scanning the street with a frantic energy that had nothing to do with the weather."
1
"A string of emergency lights cast a sickly yellow glow down a long, curved platform, illuminating rails that disappeared into an abyss of black."
2
"The only other sound was the steady drip, drip, drip from above and the low, resonant hum that seemed to vibrate up through the soles of her shoes."
3
"Cages held creatures that skittered and glowed, and jars on shelves contained swirling nebulae of coloured dust."
100.00%
Useless dialogue additions
Target: ≤5% dialogue tags with trailing filler fragments
totalTags
0
uselessAdditionCount
0
matches
(empty)
100.00%
Dialogue tag variety (said vs. fancy)
Target: ≤10% fancy dialogue tags
totalTags
0
81.0719%