NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-17-01KHMR8QBQATJY920Q1219MJRY
Run Details
Model:
Z.AI GLM 4.6
Test:
Bad Writing Habits
Scenario:
Thriller: chase through city streets
Duration:
33.0s
Cost:
$0.007484
Provider:
Z.AI
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 lashed the streets of Soho, turning the pavement into a slick, black mirror that
fractured
the glare of headlights and neon signs. Detective Harlow Quinn’s lungs burned, a familiar fire that did little to slow the relentless, rhythmic pump of her legs. Ahead of her, the figure of Tomás Herrera was a fleeting shadow, his dark curls plastered to his skull, the worn leather of his jacket darkening with every second he spent in the downpour. He was fast, she’d give him that. Not the trained, efficient speed of a professional, but the panicked, desperate scramble of a cornered animal. He vaulted over a overflowing bin, sending a cascade of refuse onto the pavement, and Quinn followed with a economy of motion that bordered on instinct. Her 41 years felt like 21 in moments like this, the muscle memory of eighteen years on the force taking over. She ignored the cold seeping through her coat, the water dripping from her closely cropped salt-and-pepper hair into her eyes. Her focus was a pinpoint, locked onto the back of Herrera’s neck. He’d slipped out of The Raven’s Nest just as she’d been closing in. She’d been
watch
ing the place for a week, certain the bar’s quiet owner, Silas, was a hub for the strange little clique she’d been tracking. Herrera was a known associate, an off-the-books medic with a history of treating patients who didn’t, couldn’t, exist in any official NHS database. The moment he’d seen her unmarked car, he’d bolted. “Police! Stop!” she’d yelled, the command swallowed by the storm. Useless. He cut left down a narrow, gas-lit alley that stank of damp brick and overflowing skips. Quinn’s worn leather
watch
on her left wrist slapped against her bone as she pumped her arms, the only sound besides their ragged breathing and the drumming of the rain. She saw him stumble, a sharp cry swallowed by the storm, and for a fleeting second, she thought she had him. He’d gone down on one knee, his hand flying to his left forearm, to the scar she knew was there from a knife fight years ago. But he was up again in a flash, his Saint Christopher medallion a brief, golden
glint
against his olive skin before it disappeared under his collar. The alley opened onto a wider street, quieter, lined with shuttered shops. This was Camden. He
was leading
her away from his territory, or maybe toward it. She couldn’t be sure. The chase had become a puzzle. He wasn’t just running; he
was navigating
. He dodged across the road, forcing a taxi to slam on its brakes, its horn blaring a useless, angry protest into the night. Quinn didn’t hesitate, her gaze fixed on her quarry. He
was heading
for the Tube station. A dead end at this hour. Trains
weren’t running
. He
was cornered
. But Herrera didn’t run for the main entrance. He skirted the building, his feet splashing through deep puddles that collected in the shadows, and disappeared down a steep, unmarked decline. A service entrance. Abandoned. Quinn followed, her hand instinctively dropping to rest on the grip of her sidearm. The air changed here. It grew colder, thick with the smell of damp earth, rot, and something else. Something acrid and metallic, like old blood and ozone. The sound of the rain faded to a distant drumming, replaced by the drip-drip-drip of water from the arched brick ceiling above. The passage opened into a small, tiled area. Faded posters for long-defunct musicals peeled from the walls. It was the disused entrance to an abandoned part of the station. Herrera
was standing
before a set of heavy steel doors, his back to her, his shoulders heaving. He
wasn’t trying
to force them open. He
was fumbling
with something in his pocket. “Herrera,” Quinn said, her voice sharp and clear in the enclosed space. She had him. “It’s over.” He turned
slowly
. His warm brown eyes, usually so calm and reassuring in the photos she’d seen, were wide with a mixture of terror and defiance. Rainwater streamed down his face. In his hand, he held a small, carved object. It looked like bone. “You don’t want to do this, Detective,” he said, his voice raspy. He had a Spanish accent, thicker now with stress. “You need to turn around.” “I’m not turning around,” she said, taking a step forward. “You’re coming with me. We have a lot to talk about. The Raven’s Nest. Your patients. Silas.” He shook his head, a sad, almost pitying gesture. “You don’t understand anything.” He held up the bone token. It was intricately carved, a series of interlocking spirals. As he pressed it against a small, dark recess on the steel door, a low hum vibrated through the floor, up Quinn’s legs. There was no click of a lock. The air simply… parted. A seam of light, sickly and green, appeared in the center of the door, widening into a vertical slit that wasn’t a door at all, but a tear in the world. A wave of air hit her, carrying a
cacophony
of smells and sounds. Spices she couldn’t name, the sweet cloy of burnt sugar, the sharp tang of chemicals, and beneath it all, a low, chattering murmur of dozens of voices. It was the sound of a marketplace, but one that existed in the gutters of a nightmare. Herrera didn’t hesitate. He stepped through the tear of light and vanished. Quinn froze, her training screaming at her. This was impossible. This was the kind of thing that had gotten her partner, DS Morris, killed. Unexplained. Supernatural. The official report had cited a freak gas explosion, a structural collapse. But Harlow had been there. She’d seen the impossible geometry of the room, felt the cold that wasn’t just a lack of heat. She’d seen the look in Morris’s eyes right before… before. They’d called her traumatized. Grief-stricken. They’d
gently
sidelined her, passed her over for promotion, until she’d become a ghost in her own department, chasing whispers and rumors on her own time. And here it was again. The Veil. The Market. The thing she’d only read about in
fragmented
, redacted reports she wasn’t supposed to have access to. The tear in the door
began to
shrink, the green light pulsing like a dying heart. Herrera
was gone
. The evidence
was gone
. Her entire case, built on months of painstaking surveillance, was about to literally disappear into another reality. She could call it in. She could report a suspect fled into a secured area. They’d send a unit, they’d find a set of locked, rusted doors. They’d think she was crazy, or worse, that she
was lying
to cover up her failure. They’d take her badge. For good this time. Or she could follow. She thought of Morris, of his easy smile, of the way he’d bought her coffee on cold mornings. He’d walked into a room like this, and he’d never walked out. Fear, cold and sharp, coiled in her gut. But beneath it, something else stirred. A grim, relentless curiosity. A need to know. A need for answers that had festered for three years. The tear was now no wider than her hand. The sounds from within
were fading
, the light dimming. Quinn didn’t remember making a decision. She only remembered moving. Her body acted on an impulse she didn’t know she possessed. She took three long strides, covering the distance to the door, and plunged her hand into the shrinking, green-
tinged
wound in the world. It felt like plunging her arm into ice water and static electricity at the same time. The pull was immense. She braced herself, closed her eyes, and stepped through after him.
created_at
"2026-02-17T02:54:02.118887+00:00"
2/17/2026, 2:54:02 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
1170
moderation_latency
(null)
generation_time
32898
tokens_prompt
914
tokens_completion
3756
native_tokens_prompt
813
native_tokens_completion
3361
native_tokens_completion_images
(null)
native_tokens_reasoning
1679
native_tokens_cached
812
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
0.00748412
router
(null)
provider_responses
0
id
"20260217105328a357e3c73f654af4"
status
200
is_byok
false
latency
1170
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-1771296808-cZfACgHWz8ukGsX8RKvB"
upstream_id
"20260217105328a357e3c73f654af4"
total_cost
0.00748412
cache_discount
0.00039788
upstream_inference_cost
0
provider_name
"Z.AI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
3
adverbTagCount
0
adverbTags
(empty)
84.5798%
AI-ism adverb frequency
Target: <1% AI-ism adverbs (57 tracked)
wordCount
1297
totalAiIsmAdverbs
2
found
0
adverb
"gently"
count
1
1
adverb
"slowly"
count
1
highlights
0
"gently"
1
"slowly"
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)
53.7394%
AI-ism word frequency
Target: <1% AI-ism words (140 tracked)
wordCount
1297
totalAiIsms
6
found
0
word
"cacophony"
count
1
1
word
"tinged"
count
1
2
word
"glint"
count
1
3
word
"fragmented"
count
1
4
word
"fractured"
count
1
5
word
"navigating"
count
1
highlights
0
"cacophony"
1
"tinged"
2
"glint"
3
"fragmented"
4
"fractured"
5
"navigating"
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
114
matches
(empty)
100.0000%
Filter word density
Target: ≤12% sentences with filter/hedge words
filterCount
1
hedgeCount
1
narrationSentences
114
filterMatches
0
"watch"
hedgeMatches
0
"began to"
100.0000%
Overuse of "that" (subordinate clause padding)
Target: ≤10% sentences with "that" clauses
thatCount
0
totalSentences
117
matches
(empty)
100.0000%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
26
mean
49.5
std
29.01
cv
0.586
sampleLengths
0
77
1
103
2
70
3
11
4
119
5
44
6
51
7
34
8
63
9
61
10
17
11
44
12
26
13
27
14
51
15
42
16
57
17
12
18
102
19
26
20
40
21
51
22
4
23
62
24
18
25
75
100.0000%
Passive voice overuse
Target: ≤5% passive sentences
passiveCount
3
totalSentences
114
matches
0
"was cornered"
1
"was gone"
2
"was gone"
100.0000%
Past progressive (was/were + -ing) overuse
Target: ≤10% past progressive verbs
pastProgressiveCount
9
totalVerbs
206
matches
0
"was leading"
1
"was navigating"
2
"was heading"
3
"weren’t running"
4
"was standing"
5
"wasn’t trying"
6
"was fumbling"
7
"was lying"
8
"were fading"
94.0000%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
1255
adjectiveStacks
1
stackExamples
0
"narrow, gas-lit alley"
adverbCount
28
adverbRatio
0.022310756972111555
lyAdverbCount
12
lyAdverbRatio
0.009561752988047808
100.0000%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
117
echoCount
0
echoWords
(empty)
100.0000%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
117
mean
11
std
7.85
cv
0.713
sampleLengths
0
23
1
20
2
34
3
7
4
17
5
26
6
21
7
20
8
12
9
13
10
25
11
23
12
9
13
10
14
1
15
16
16
30
17
21
18
26
19
26
20
12
21
3
22
12
23
4
24
6
25
7
26
23
27
9
28
7
29
6
30
3
31
3
32
8
33
22
34
3
35
1
36
14
37
4
38
14
39
9
40
22
41
8
42
9
43
12
44
17
45
7
46
8
47
12
48
3
49
2
41.3105%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
15
diversityRatio
0.3247863247863248
totalSentences
117
uniqueOpeners
38
8.3333%
Dialogue tag variety (said vs. fancy)
Target: ≤30% fancy dialogue tags
totalTags
3
fancyCount
2
fancyTags
0
"Quinn said, her voice sharp and clear in the enclosed space (say, clear enclose)"
1
"she said, taking a step forward (say, take)"
87.1743%