NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-02-17-01KHMQKMD7TRV9H0VA98MEXW72
Run Details
Model:
Z.AI GLM 4.6
Test:
Bad Writing Habits
Scenario:
Thriller: chase through city streets
Duration:
40.8s
Cost:
$0.008348
Provider:
Z.AI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Adverbs in dialogue tags
(details)
89.97%
AI-ism adverb frequency
(details)
100.00%
AI-ism character names
(details)
100.00%
AI-ism location names
(details)
39.80%
AI-ism word frequency
(details)
100.00%
Cliché density
(details)
100.00%
Emotion telling (show vs. tell)
(details)
87.91%
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)
15.59%
Narrator intent-glossing
(details)
0.00%
"Not X but Y" pattern overuse
(details)
100.00%
Overuse of "that" (subordinate clause padding)
(details)
100.00%
Paragraph length variance
(details)
97.17%
Passive voice overuse
(details)
32.64%
Past progressive (was/were + -ing) overuse
(details)
54.95%
Em-dash & semicolon overuse
(details)
100.00%
Purple prose (modifier overload)
(details)
100.00%
Repeated phrase echo
(details)
100.00%
Sentence length variance
(details)
33.08%
Sentence opener variety
(details)
27.10%
Adverb-first sentence starts
(details)
57.40%
Pronoun-first sentence starts
(details)
37.24%
Subject-first sentence starts
(details)
0.00%
Subordinate conjunction sentence starts
(details)
80.75%
Technical jargon density
(details)
100.00%
Useless dialogue additions
(details)
100.00%
Dialogue tag variety (said vs. fancy)
(details)
75.1191%
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
The rain fell in relentless, greasy sheets, turning the London night into a smear of neon and asphalt. Detective Harlow Quinn’s lungs burned, a
familiar
fire that did nothing to slow the piston-pound of her feet against the slick pavement. Fifty yards ahead, the suspect—a man in a dark jacket with short, curly hair—ducked around the corner of a brick office block.
He moved with an unnerving
fluidity,
not like a panicked criminal but like a man who knew the city’s secret veins
.
Quinn followed, her worn leather
watch
a cold,
familiar
weight
on her left wrist.
Her service weapon was a
reassuring pressure at her hip, but she knew this chase wouldn’t end with a clean shot. This one was different.
The case file, thin on
facts and thick on weirdness, had the same stink as the one that took DS Morris three years ago.
A feeling of wrongness that
clung to the edges of reality.
She rounded the corner into
a narrower alley, the air thick with the smell of damp rubbish and stagnant water.
The suspect was already halfway
down, his silhouette framed by the distant orange glow of a streetlamp.
He glanced back, and for
a split second, their eyes met across the rain-lashed distance.
Quinn saw no fear, only
a
calculated
assessment.
Then he vaulted over a
locked metal gate with an ease that defied his build. Cursing under her breath, Quinn put on a burst of speed. She was forty-one, not twenty, but her body, honed by years of military-precise discipline, responded.
She hit the gate at
a run, scrambled up, and swung her leg over, the cold metal biting into her palms.
She landed hard, jarring her
knees, but didn’t break stride.
The alley opened out into
the
chaotic
thrum of Camden.
The streets were a river
of people, umbrellas bobbing like dark flowers. The suspect was a ghost, melting into the crowd. Quinn’s sharp jaw tightened. She couldn’t lose him.
He was the first solid
lead in weeks, a paramedic-turned-back-alley-sawbones who’d shown up on CCTV near two impossible crime scenes. Tomás Herrera. That was the name on the file. She pushed through the throng, her gaze fixed on the dark blue of his jacket.
He wasn’t running anymore, just
walking with purpose, weaving through the late-night shoppers and pub-goers. He
was leading
her somewhere. The thought sent a prickle of unease up her spine. This wasn’t a flight; it was an escort.
He turned down a side
street, away from the noise and light, heading toward a quieter, more industrial part of the borough. The streetlights here were fewer, casting long, distorted shadows. The only sound was the drumming of the rain and the
echo
of their
footsteps
—his steady, hers a ragged counterpoint.
He stopped before a set
of grimy concrete stairs leading down to what looked like a disused Tube station entrance. A faded sign, its letters peeled away by time, was barely legible. Above it, crude graffiti covered the tiled walls. This was it. A dead end. Or a beginning. Quinn slowed, pressing herself into the shadows of a loading bay across the street, her hand resting on the grip of her weapon.
She watched, her breath misting
in the cold air. Herrera didn’t hesitate.
He descended the stairs, his
movements sure.
He stopped before a heavy
steel door, one that looked like it hadn’t
been opened
in decades.
He reached into his jacket
and pulled out something small and pale.
He held it up to
the door. From this distance, Quinn couldn’t see what it was, but it caught the faint light—a sliver of bone. There was a low click, a grating sound of old locks disengaging, and the door swung inward, spilling
not darkness, but a strange, warm, amber light onto the wet concrete
. A murmur of voices, a scent of incense and ozone, drifted up to her. Her training screamed at her. *Call for backup. Secure the perimeter. Do not enter alone.* Every instinct honed over eighteen years on the force told her this was a breach of protocol, a one-way trip into a world she didn’t understand. This was how Morris had died. Chasing whispers into the dark. But the memory of Morris’s face, pale and still in a morgue that couldn’t explain his wounds, was the
very
thing that propelled her forward. This was her second chance to stare into the abyss and find out what
was looking
back. She couldn’t let it go. Not again. The door clicked shut behind Herrera, plunging the entrance back into shadow. Quinn waited a count of ten, then crossed the street, her
footsteps
silent on the wet pavement. She stood before the steel door, the faint, humming energy of it vibrating through the soles of her boots. There was no handle, no visible lock. The bone token. It was a key. And she didn’t have one. Frustration warred with a grim, desperate curiosity. She ran her hands over the cold steel, looking for a weakness, a seam. Nothing. It was solid. She was about to step back, to reconsider, when she heard another sound from below. The grating click of the lock again. The door
began to
swing open. Quinn flattened herself against the wall, her heart hammering against her ribs. Two figures emerged, cloaked and hooded, their conversation a low, guttural language she didn’t recognize. They moved past her without a glance, disappearing into the London night. The door remained ajar, a sliver of that impossible light beckoning to her. This was her chance. She didn’t let herself
think
. She slipped through the opening, pulling the heavy door quietly shut behind her. The air hit her like a physical blow. It was thick, heavy with the smells of spiced smoke, damp earth, and something metallic, like old coins. The narrow tunnel opened up into a cavernous space. An abandoned Tube station, just as the file had suggested. The curved platform was still there, but the tracks
were gone
, replaced by a sprawling,
chaotic
market. The Veil Market. It was a
kaleidoscope
of impossible sights. Stalls
were crammed
together under flickering bare bulbs and the soft glow of what
looked like trapped lightning in jars
. A woman with silver hair sold bottled whispers that
shimmered
like heat haze. A
hulking
man with a tattooed face displayed
gleaming
, serrated
blades that seemed to drink the light
. At another stall, a cloaked figure exchanged a small pouch of coins for a folded piece of paper, which promptly burst into blue flame and vanished. The low murmur of a dozen different languages filled the air, punctuated by the chime of strange bells and the occasional hiss of a released alchemical vapor. Quinn felt like a
predator
in a world of mythical beasts. Her police-issue raincoat and practical boots were a uniform from another reality. She kept to the shadows of the tiled walls, her cop’s eyes
scanning
the crowd, her movements economical and silent. She
was looking
for one face in this sea of strangers. And then she saw him. Herrera. He was deeper in the market, standing at a stall piled high with herbs and dried fungi. He’d taken off his jacket, and the simple white t-shirt he wore did little to hide the lean, capable lines of his body. The Saint Christopher medallion around his neck glinted
softly
. He
wasn't just buying
something. He
was talking
to the stall keeper, an old woman with eyes like shards of obsidian, his hands moving as he described something with intense focus. He looked
completely
at home. This was his world. Quinn’s hand tightened on her weapon. She could make the arrest now. Plow through the crowd, identify herself, and cuff him. It would be a mess. A violent,
chaotic
mess in a place where the rules didn’t apply. They would tear her apart. Or she could wait. Watch. Learn. Her gaze fell on the scar that ran along his left forearm, a pale, jagged line against his olive skin. A knife attack, the file said. He knew violence. He wasn’t just some back-alley doctor. He was a survivor. He finished his transaction, tucking a small, wrapped parcel into his jacket pocket. He turned, his warm brown eyes
scanning
the crowd, and for a heart-stopping second, Quinn thought he saw her. But his gaze slid past, and he started moving, heading toward a darker, less-traveled tunnel at the far end of the platform. The decision
crystallized
in her mind, sharp and cold as broken glass. Backup would never get here in time. A report would get her laughed out of the precinct and buried under paperwork. This was her case. Her ghost. She let out a slow, silent breath, the air tasting of magic and danger. She pushed off the wall and
began to
follow, a shadow in a world of shadows, descending deeper into the London that didn't exist on any map.
created_at
"2026-02-17T02:42:38.718382+00:00"
2/17/2026, 2:42:38 AM
model
"z-ai/glm-4.6"
4/6/2001, 12:00:00 AM
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1404
moderation_latency
(null)
generation_time
40732
tokens_prompt
914
tokens_completion
4050
native_tokens_prompt
813
native_tokens_completion
3574
native_tokens_completion_images
(null)
native_tokens_reasoning
1681
native_tokens_cached
6
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
0.00834766
router
(null)
provider_responses
0
id
"2026021710415775a44ae4e8084f87"
status
200
is_byok
false
latency
1404
endpoint_id
"a87bfa67-bad9-4ce3-8c20-99b61249cfc3"
provider_name
"Z.AI"
model_permaslug
"z-ai/glm-4.6"
4/6/2001, 12:00:00 AM
api_type
"completions"
id
"gen-1771296117-gbIXtrGzHmyW9bs1yDEM"
upstream_id
"2026021710415775a44ae4e8084f87"
total_cost
0.00834766
cache_discount
0.00000294
upstream_inference_cost
0
provider_name
"Z.AI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
0
89.97%
AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount
1495
totalAiIsmAdverbs
3
found
0
adverb
"very"
count
1
1
adverb
"softly"
count
1
2
adverb
"completely"
count
1
highlights
0
"very"
1
"softly"
2
"completely"
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)
39.80%
AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount
1495
totalAiIsms
18
found
0
word
"familiar"
count
2
1
word
"weight"
count
1
2
word
"calculated"
count
1
3
word
"chaotic"
count
3
4
word
"echo"
count
1
5
word
"footsteps"
count
2
6
word
"kaleidoscope"
count
1
7
word
"shimmered"
count
1
8
word
"hulking"
count
1
9
word
"gleaming"
count
1
10
word
"predator"
count
1
11
word
"scanning"
count
2
12
word
"crystallized"
count
1
highlights
0
"familiar"
1
"weight"
2
"calculated"
3
"chaotic"
4
"echo"
5
"footsteps"
6
"kaleidoscope"
7
"shimmered"
8
"hulking"
9
"gleaming"
10
"predator"
11
"scanning"
12
"crystallized"
100.00%
Cliché density
Target: ≤1 cliche(s) per 800-word window
totalCliches
0
maxInWindow
0
found
(empty)
highlights
(empty)
100.00%
Emotion telling (show vs. tell)
Target: ≤3% sentences with emotion telling
emotionTells
0
narrationSentences
130
matches
(empty)
87.91%
Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount
2
hedgeCount
3
narrationSentences
130
filterMatches
0
"watch"
1
"think"
hedgeMatches
0
"began to"
1
"seemed to"
100.00%
Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences
130
gibberishSentences
0
adjustedGibberishSentences
0
longSentenceCount
0
runOnParagraphCount
0
giantParagraphCount
0
wordSaladCount
0
repetitionLoopCount
0
controlTokenCount
0
maxSentenceWordsSeen
31
ratio
0
matches
(empty)
100.00%
Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans
1
markdownWords
10
totalWords
1491
ratio
0.007
matches
0
"Call for backup. Secure the perimeter. Do not enter alone."
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
30
wordCount
1491
uniqueNames
11
maxNameDensity
0.8
worstName
"Quinn"
maxWindowNameDensity
1.5
worstWindowName
"Quinn"
discoveredNames
London
3
Harlow
1
Quinn
12
Morris
3
Camden
1
Herrera
4
Tube
2
Veil
1
Market
1
Saint
1
Christopher
1
persons
0
"Harlow"
1
"Quinn"
2
"Morris"
3
"Herrera"
4
"Saint"
5
"Christopher"
places
0
"London"
globalScore
1
windowScore
1
15.59%
Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences
93
glossingSentenceCount
5
matches
0
"looked like a disused Tube station entran"
1
"looked like it hadn’t been opened in deca"
2
"looked like trapped lightning in jars"
3
"blades that seemed to drink the light"
4
"felt like a predator in a world of myth"
0.00%
"Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches
3
per1kWords
2.012
wordCount
1491
matches
0
"not like a panicked criminal but like a man who knew the city’s secret veins"
1
"not twenty, but her body, honed by years of military-precise discipline, res"
2
"not darkness, but a strange, warm, amber light onto the wet concrete"
100.00%
Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount
0
totalSentences
130
matches
(empty)
100.00%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
25
mean
59.64
std
35.94
cv
0.603
sampleLengths
0
83
1
74
2
132
3
83
4
38
5
91
6
9
7
140
8
52
9
49
10
67
11
53
12
52
13
22
14
62
15
3
16
110
17
54
18
6
19
89
20
43
21
6
22
39
23
54
24
80
97.17%
Passive voice overuse
Target: ≤2% passive sentences
passiveCount
3
totalSentences
130
matches
0
"been opened"
1
"were gone"
2
"were crammed"
32.64%
Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount
6
totalVerbs
239
matches
0
"wasn’t running"
1
"was leading"
2
"was looking"
3
"was looking"
4
"wasn't just buying"
5
"was talking"
54.95%
Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount
4
semicolonCount
1
flaggedSentences
4
totalSentences
130
ratio
0.031
matches
0
"Fifty yards ahead, the suspect—a man in a dark jacket with short, curly hair—ducked around the corner of a brick office block."
1
"This wasn’t a flight; it was an escort."
2
"The only sound was the drumming of the rain and the echo of their footsteps—his steady, hers a ragged counterpoint."
3
"From this distance, Quinn couldn’t see what it was, but it caught the faint light—a sliver of bone."
100.00%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
1510
adjectiveStacks
0
stackExamples
(empty)
adverbCount
40
adverbRatio
0.026490066225165563
lyAdverbCount
9
lyAdverbRatio
0.005960264900662252
100.00%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
130
echoCount
0
echoWords
(empty)
100.00%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
130
mean
11.47
std
7.25
cv
0.632
sampleLengths
0
18
1
22
2
22
3
21
4
14
5
21
6
4
7
24
8
11
9
20
10
17
11
15
12
8
13
15
14
11
15
15
16
21
17
10
18
10
19
12
20
9
21
4
22
4
23
20
24
2
25
7
26
15
27
15
28
5
29
10
30
8
31
22
32
9
33
20
34
20
35
12
36
8
37
3
38
3
39
3
40
23
41
9
42
3
43
7
44
17
45
12
46
7
47
18
48
30
49
14
33.08%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
22
diversityRatio
0.2846153846153846
totalSentences
130
uniqueOpeners
37
27.10%
Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount
1
totalSentences
123
matches
0
"Then he vaulted over a"
ratio
0.008
57.40%
Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount
50
totalSentences
123
matches
0
"He moved with an unnerving"
1
"Her service weapon was a"
2
"She rounded the corner into"
3
"He glanced back, and for"
4
"She was forty-one, not twenty,"
5
"She hit the gate at"
6
"She landed hard, jarring her"
7
"She couldn’t lose him."
8
"He was the first solid"
9
"She pushed through the throng,"
10
"He wasn’t running anymore, just"
11
"He was leading her somewhere."
12
"He turned down a side"
13
"He stopped before a set"
14
"She watched, her breath misting"
15
"He descended the stairs, his"
16
"He stopped before a heavy"
17
"He reached into his jacket"
18
"He held it up to"
19
"Her training screamed at her."
ratio
0.407
37.24%
Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount
104
totalSentences
123
matches
0
"The rain fell in relentless,"
1
"Detective Harlow Quinn’s lungs burned,"
2
"He moved with an unnerving"
3
"Quinn followed, her worn leather"
4
"Her service weapon was a"
5
"This one was different."
6
"The case file, thin on"
7
"A feeling of wrongness that"
8
"She rounded the corner into"
9
"The suspect was already halfway"
10
"He glanced back, and for"
11
"Quinn saw no fear, only"
12
"She was forty-one, not twenty,"
13
"She hit the gate at"
14
"She landed hard, jarring her"
15
"The alley opened out into"
16
"The streets were a river"
17
"The suspect was a ghost,"
18
"Quinn’s sharp jaw tightened."
19
"She couldn’t lose him."
ratio
0.846
0.00%
Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount
0
totalSentences
123
matches
(empty)
ratio
0
80.75%
Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences
69
technicalSentenceCount
6
matches
0
"Detective Harlow Quinn’s lungs burned, a familiar fire that did nothing to slow the piston-pound of her feet against the slick pavement."
1
"He moved with an unnerving fluidity, not like a panicked criminal but like a man who knew the city’s secret veins."
2
"A feeling of wrongness that clung to the edges of reality."
3
"He was the first solid lead in weeks, a paramedic-turned-back-alley-sawbones who’d shown up on CCTV near two impossible crime scenes."
4
"Quinn slowed, pressing herself into the shadows of a loading bay across the street, her hand resting on the grip of her weapon."
5
"Her gaze fell on the scar that ran along his left forearm, a pale, jagged line against his olive skin."
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
75.1191%