NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-02-17-01KHMQQXN2WR5127T3C99XGD7G
Run Details
Model:
Z.AI GLM 4.6
Test:
Bad Writing Habits
Scenario:
Literary fiction: old friends reunite
Duration:
1m 3s
Cost:
$0.006793
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
100.00%
Adverbs in dialogue tags
(details)
86.60%
AI-ism adverb frequency
(details)
100.00%
AI-ism character names
(details)
100.00%
AI-ism location names
(details)
36.33%
AI-ism word frequency
(details)
33.33%
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)
58.54%
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)
95.04%
Passive voice overuse
(details)
38.19%
Past progressive (was/were + -ing) overuse
(details)
92.29%
Em-dash & semicolon overuse
(details)
94.00%
Purple prose (modifier overload)
(details)
100.00%
Repeated phrase echo
(details)
100.00%
Sentence length variance
(details)
41.30%
Sentence opener variety
(details)
0.00%
Adverb-first sentence starts
(details)
13.81%
Pronoun-first sentence starts
(details)
16.70%
Subject-first sentence starts
(details)
51.55%
Subordinate conjunction sentence starts
(details)
100.00%
Technical jargon density
(details)
0.00%
Useless dialogue additions
(details)
100.00%
Dialogue tag variety (said vs. fancy)
(details)
75.2557%
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="4412" name="Aurora Carter" aliases="Rory, Laila, Aurora, Carter, Malphora" pronouns="she/her" species="Human" storyRole="Protagonist"> <eyeColor>bright blue</eyeColor> <hair>straight shoulder-length black</hair> <height>5'6"</height> <age>25</age> <scar>small crescent-shaped scar on her left wrist from a childhood accident</scar> <description>Aurora, referred to by her friends as Rory, is cool-headed, intelligent, and known for quick out-of-the-box thinking.</description> <backstory>Born in Cardiff to an Irish barrister father (Brendan Carter) and Welsh teacher mother (Jennifer Carter née Ellis), Rory reluctantly pursued Pre-Law at Cardiff University. She fled to London at the behest of her childhood friend Eva after leaving an abusive ex named Evan. By day, Rory works part-time as a delivery person for Yu-Fei Cheung's Golden Empress restaurant. She lives in a flat above Silas' bar.</backstory> </character> <character id="4413" name="Silas Blackwood" aliases="Silas, Si, the Spymaster" pronouns="he/him" species="Human" storyRole="Mentor"> <eyeColor>hazel</eyeColor> <hair>grey-streaked auburn</hair> <height>6'1"</height> <age>58</age> <beard>neatly trimmed, matching his grey-streaked auburn hair</beard> <limp>slight limp in his left leg from an old knee injury</limp> <signetRing>silver signet ring, always worn on his right hand</signetRing> <description>Silas is a retired intelligence operative turned bar owner. He carries himself with quiet authority.</description> <backstory>Former MI6 field agent who retired after a botched operation in Prague left him with a permanent knee injury. He opened "The Raven's Nest" bar in Soho as a front for his network of contacts.</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> </codex> Write a fully fleshed out scene of at least 1000 words in the literary fiction genre. Two old friends who haven't seen each other in years run into each other unexpectedly at a bar. One of them has changed significantly. Through their conversation and the tension between who they were and who they've become, reveal the weight of time and unspoken regret. 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 bell above the door
gave a tired jangle as Rory pushed her way into The Raven’s Nest.
The air, thick with the
ghosts of spilled beer and lemon polish, settled over her like a
familiar
coat. Outside, the Soho streets were a riot of Friday night noise, but in here, the world was muted, held at bay by dark wood and dim lighting.
The distinctive green neon sign
cast a watery glow through the front window, painting the tables in shades of underwater algae.
It was her sanctuary, the
place she lived above, the place she worked to forget.
She wasn't on duty for
Yu-Fei tonight, but the ten-hour shift delivering aromatic cartons of Mapo Tofu and Sichuan chicken across London had left a grit in her eyes and a dull ache in her shoulders. All she wanted was a glass of water and the
solitary
comfort of a corner table. But he wasn't behind the bar.
That was the first thing
that felt wrong. Silas was a fixture there, a quiet monolith polishing glasses or dispensing gruff wisdom with the economy of a poet.
His absence left a gap
in the room's geometry. Rory’s gaze swept the space, past the walls papered in faded maps of Cold War-era Europe and black-and-white photographs of men in trench coats who looked far too serious for their own good.
She found him in a
booth near the back, half-swallowed by shadow. He
wasn't holding
court.
He wasn't watching the door
with the assessing gaze of the retired spymaster he was. He was just… sitting. Staring into a half-empty glass of what
looked like whiskey
.
The sight of him sent
a jolt through her, something colder than the autumn
chill
outside.
It had been a few
weeks since they’d had a proper talk. She’d seen him, of course. A nod on the stairs, a brief "alright?" across the bar. But this was different.
This was stillness, and Silas
was never still.
She slid onto the bench
opposite him.
The old vinyl groaned under
her
weight
. "Rough night?"
His hazel eyes, usually so
sharp and clear,
were clouded
, like river stones smoothed and obscured by a current.
It took a moment for
them to focus on her.
A flicker of recognition
, then a slow, tired smile that didn’t
quite reach his eyes
. "Rory. Didn't hear you come in."
His voice was lower than
usual, gravelly. "I'm stealthy,"
she said, trying for a
lightness she didn't feel. "It's in the job description. Part-time delivery ninja."
She gestured to his glass
. "Celebrating?" He huffed a quiet laugh, a sound like shifting stones. "Something like that."
He looked down at his
hand, resting on the table. The silver signet ring he always wore
seemed to
catch the meagre light, a small, defiant spark. But his knuckles, Rory noticed, were swollen, the skin stretched thin over the joints. He looked older. The grey in his neatly trimmed auburn beard wasn't just distinguished anymore; it was a declaration of defeat. The lines
etched
around his mouth weren't from laughter; they were from holding things in. "It's been a while,"
she said softly
. The words felt inadequate, a flimsy raft on a deep, dark ocean. "Has it?"
He lifted his glass and
swirled the amber liquid. "Time gets… compressed. One day bleeds into the next. You wake up and you're twenty years older and you wonder where the hell the interval went." She didn't know what to say to that. This wasn't the Silas who had found her a job, who had given her a flat when she had nowhere else to go, the man who carried himself with the quiet authority of a king in exile. This was a man stripped of his armour, and the sight of it was profoundly unsettling. She thought of the man she used to know, the one who could read her with a glance, whose advice was always sharp, practical, and unerringly right. The man who had looked at her, a terrified refugee from her own life, and simply said, "The past is a place you visit, not where you live." "Are you okay, Si?" The question
hung in the air
between them, fragile and dangerous. He finally looked at her,
really
looked at her, and for a second, she saw the old
intensity
flash in his eyes. "I'm an old man with a bad knee, Rory. What's 'okay' got to do with it?"
As if to prove the
point, he shifted his left leg, and she heard the faint, grating friction that was the soundtrack to his every move. The legacy of Prague. "My dad's getting old," she said, the words tumbling out, a clumsy attempt to bridge the chasm. "He complains about his back. He says it's the price of being a barrister, sitting in uncomfortable chairs for thirty years." "Your father,"
Silas said, a ghost of a smile on his lips
. "How is Brendan? Still convinced you're going to go back to Cardiff and finish your Pre-Law degree?" The mention of her abandoned life, the one she'd fled with Eva's help, landed like a stone in her gut. That was another lifetime. The girl who was going to be a lawyer, who had let a man named Evan chip away at her until she was a hollowed-out version of herself. She twisted a strand of her straight black hair, her fingers brushing unconsciously against the small crescent-shaped scar on her left wrist. A childhood accident, she always said. A lie she told so often she almost believed it. "No. I think he's given up on that," she said. "He just wants me to be happy, I think. In his own, barristerly way." "And are you?"
Silas asked, his gaze direct, piercing
. "Happy?" The question was so simple, yet it
carried the weight
of their entire history. She was safe. She was free. She
was working
, she had a roof over her head. But happy? That felt like a language she'd forgotten how to speak. "I'm not unhappy," she settled on, which was the truth. It was a neutral, livable state. Silas nodded
slowly
, as if that was the most
profound
thing he'd ever heard. He took a sip of his whiskey. "Good enough,"
he murmured, more to himself than to her
. "Sometimes, good enough is all you get." A heavy
silence
descended, filled by the low murmur of other patrons and the clink of ice against glass. Rory looked past him, at a photograph of a group of men standing on a Prague street in 1968. They
were smiling
, but their eyes
were haunted
. She wondered if Silas knew them. She wondered if he saw himself in them, young and full of purpose, before a botched operation and a
shattered
knee had brought him here, to this dimly lit bar in Soho, a king ruling over a small, quiet kingdom of ghosts. "You saved me, you know,"
she said, the words barely a whisper
. He didn't dismiss it with a wave of his hand. He didn't say it was nothing. He just looked at her, and in the clouded
depths
of his hazel eyes, she saw it: a
flicker
of the same regret she felt. A regret for paths not taken, for mistakes made, for the inexorable march of time that wore down even the strongest foundations. He had saved her, yes. But at what cost to him? Did he
see
her as a success story, or a constant
reminder
of the world he'd lost, the world of rescues and
clandestine
meetings in rooms hidden behind bookshelves? "I gave you a key, Rory,"
he said, his voice rough with emotion
. "You opened the door." He rubbed his signet ring with his thumb, a nervous gesture she'd never seen before. "Don't ever stand in your own way. That's the only cage that matters." The
unspoken
things filled the space between them: his failed mission, her abusive ex, his stolen future, her reclaimed present. They were two people bound together by the wreckage of their past lives, sitting in a bar that was a monument to secrets. The
weight
of it was immense, but it
wasn't crushing
. It was… solid. Real. He pushed his glass towards her. "Here. You look like you need it more than I do." Rory wrapped her fingers around the cool, heavy glass. She didn't drink. She just held it, feeling the condensation bead on her skin. She looked at the man across from her, the mentor, the friend, the spymaster. He was tired. He
was broken
in ways she could only guess at. He had changed so much. But as he watched her, a trace of the old fire returned to his eyes, and she
knew that beneath
the weariness, the man who had given her a key was still there. And for the first time in a long time, she felt
not just safe, but truly seen
.
created_at
"2026-02-17T02:45:22.393159+00:00"
2/17/2026, 2:45:22 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
145
moderation_latency
(null)
generation_time
62796
tokens_prompt
867
tokens_completion
4155
native_tokens_prompt
820
native_tokens_completion
3856
native_tokens_completion_images
(null)
native_tokens_reasoning
2076
native_tokens_cached
768
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
0.0067932399994624
router
(null)
provider_responses
0
id
"chatcmpl-RmrgOjwUhyOEHq9cwzxZ45Hh"
status
200
is_byok
false
latency
23
endpoint_id
"c8607edc-8577-4199-b03d-1213da4743dc"
provider_name
"DeepInfra"
model_permaslug
"z-ai/glm-4.6"
4/6/2001, 12:00:00 AM
api_type
"completions"
id
"gen-1771296257-H5zmyWZNCAm7TlxEggT1"
upstream_id
"chatcmpl-RmrgOjwUhyOEHq9cwzxZ45Hh"
total_cost
0.0067932399994624
cache_discount
0.0002688000005376
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
16
adverbTagCount
1
adverbTags
0
"she said softly [softly]"
dialogueSentences
29
tagDensity
0.552
leniency
1
rawRatio
0.063
effectiveRatio
0.063
86.60%
AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount
1492
totalAiIsmAdverbs
4
found
0
adverb
"softly"
count
1
1
adverb
"really"
count
1
2
adverb
"slowly"
count
1
3
adverb
"truly"
count
1
highlights
0
"softly"
1
"really"
2
"slowly"
3
"truly"
100.00%
AI-ism character names
Target: 0 AI-default names (16 tracked, −20% each)
codexExemptions
0
"Blackwood"
found
(empty)
100.00%
AI-ism location names
Target: 0 AI-default location names (33 tracked, −20% each)
codexExemptions
(empty)
found
(empty)
36.33%
AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount
1492
totalAiIsms
19
found
0
word
"familiar"
count
1
1
word
"sanctuary"
count
1
2
word
"solitary"
count
1
3
word
"chill"
count
1
4
word
"weight"
count
2
5
word
"flicker"
count
2
6
word
"etched"
count
1
7
word
"intensity"
count
1
8
word
"carried the weight"
count
1
9
word
"profound"
count
1
10
word
"silence"
count
1
11
word
"shattered"
count
1
12
word
"whisper"
count
1
13
word
"depths"
count
1
14
word
"reminder"
count
1
15
word
"clandestine"
count
1
16
word
"unspoken"
count
1
highlights
0
"familiar"
1
"sanctuary"
2
"solitary"
3
"chill"
4
"weight"
5
"flicker"
6
"etched"
7
"intensity"
8
"carried the weight"
9
"profound"
10
"silence"
11
"shattered"
12
"whisper"
13
"depths"
14
"reminder"
15
"clandestine"
16
"unspoken"
33.33%
Cliché density
Target: ≤1 cliche(s) per 800-word window
totalCliches
3
maxInWindow
3
found
0
label
"hung in the air"
count
1
1
label
"sent a shiver through"
count
1
2
label
"flicker of emotion"
count
1
highlights
0
"hung in the air"
1
"sent a jolt through"
2
"A flicker of recognition"
100.00%
Emotion telling (show vs. tell)
Target: ≤3% sentences with emotion telling
emotionTells
0
narrationSentences
103
matches
(empty)
100.00%
Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount
1
hedgeCount
1
narrationSentences
103
filterMatches
0
"see"
hedgeMatches
0
"seemed to"
100.00%
Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences
113
gibberishSentences
0
adjustedGibberishSentences
0
longSentenceCount
0
runOnParagraphCount
0
giantParagraphCount
0
wordSaladCount
0
repetitionLoopCount
0
controlTokenCount
0
maxSentenceWordsSeen
42
ratio
0
matches
(empty)
100.00%
Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans
0
markdownWords
0
totalWords
1492
ratio
0
matches
(empty)
100.00%
Missing dialogue indicators (quotation marks)
Target: ≤10% speech attributions without quotation marks
totalAttributions
13
unquotedAttributions
0
matches
(empty)
100.00%
Name drop frequency
Target: ≤1.0 per-name mentions per 100 words
totalMentions
29
wordCount
1285
uniqueNames
17
maxNameDensity
0.54
worstName
"Silas"
maxWindowNameDensity
1.5
worstWindowName
"Silas"
discoveredNames
Rory
5
Raven
1
Nest
1
Soho
2
Friday
1
Yu-Fei
1
Mapo
1
Tofu
1
Sichuan
1
London
1
Cold
1
War-era
1
Europe
1
Silas
7
Prague
2
Eva
1
Evan
1
persons
0
"Rory"
1
"Silas"
2
"Eva"
3
"Evan"
places
0
"Raven"
1
"Soho"
2
"Sichuan"
3
"London"
4
"Europe"
5
"Prague"
globalScore
1
windowScore
1
58.54%
Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences
82
glossingSentenceCount
3
matches
0
"looked like whiskey"
1
"quite reach his eyes"
2
"felt like a language she'd forgotten ho"
100.00%
"Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches
1
per1kWords
0.67
wordCount
1492
matches
0
"not just safe, but truly seen"
100.00%
Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount
1
totalSentences
113
matches
0
"knew that beneath"
100.00%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
29
mean
51.45
std
34.03
cv
0.662
sampleLengths
0
100
1
52
2
6
3
82
4
89
5
16
6
57
7
25
8
90
9
19
10
37
11
116
12
15
13
68
14
38
15
29
16
90
17
24
18
10
19
42
20
16
21
38
22
94
23
12
24
103
25
45
26
57
27
17
28
105
95.04%
Passive voice overuse
Target: ≤2% passive sentences
passiveCount
3
totalSentences
103
matches
0
"were clouded"
1
"were haunted"
2
"was broken"
38.19%
Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount
5
totalVerbs
206
matches
0
"wasn't holding"
1
"wasn't watching"
2
"was working"
3
"were smiling"
4
"wasn't crushing"
92.29%
Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount
0
semicolonCount
2
flaggedSentences
2
totalSentences
113
ratio
0.018
matches
0
"The grey in his neatly trimmed auburn beard wasn't just distinguished anymore; it was a declaration of defeat."
1
"The lines etched around his mouth weren't from laughter; they were from holding things in."
94.00%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
1294
adjectiveStacks
1
stackExamples
0
"small crescent-shaped scar"
adverbCount
46
adverbRatio
0.03554868624420402
lyAdverbCount
15
lyAdverbRatio
0.011591962905718702
100.00%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
113
echoCount
0
echoWords
(empty)
100.00%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
113
mean
13.2
std
8.9
cv
0.674
sampleLengths
0
18
1
19
2
27
3
21
4
15
5
36
6
16
7
6
8
8
9
20
10
9
11
33
12
12
13
4
14
15
15
3
16
1
17
10
18
16
19
12
20
5
21
11
22
4
23
8
24
7
25
7
26
2
27
19
28
10
29
15
30
13
31
11
32
13
33
1
34
10
35
13
36
17
37
14
38
3
39
18
40
15
41
7
42
12
43
11
44
26
45
8
46
37
47
16
48
27
49
28
41.30%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
15
diversityRatio
0.3274336283185841
totalSentences
113
uniqueOpeners
37
0.00%
Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount
0
totalSentences
97
matches
(empty)
ratio
0
13.81%
Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount
50
totalSentences
97
matches
0
"It was her sanctuary, the"
1
"She wasn't on duty for"
2
"His absence left a gap"
3
"She found him in a"
4
"He wasn't holding court."
5
"He wasn't watching the door"
6
"He was just…"
7
"It had been a few"
8
"She’d seen him, of course."
9
"She slid onto the bench"
10
"His hazel eyes, usually so"
11
"It took a moment for"
12
"His voice was lower than"
13
"she said, trying for a"
14
"She gestured to his glass"
15
"He huffed a quiet laugh,"
16
"He looked down at his"
17
"He looked older."
18
"she said softly"
19
"He lifted his glass and"
ratio
0.515
16.70%
Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount
86
totalSentences
97
matches
0
"The bell above the door"
1
"The air, thick with the"
2
"The distinctive green neon sign"
3
"It was her sanctuary, the"
4
"She wasn't on duty for"
5
"That was the first thing"
6
"Silas was a fixture there,"
7
"His absence left a gap"
8
"Rory’s gaze swept the space,"
9
"She found him in a"
10
"He wasn't holding court."
11
"He wasn't watching the door"
12
"He was just…"
13
"The sight of him sent"
14
"It had been a few"
15
"She’d seen him, of course."
16
"A nod on the stairs,"
17
"This was stillness, and Silas"
18
"She slid onto the bench"
19
"The old vinyl groaned under"
ratio
0.887
51.55%
Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount
1
totalSentences
97
matches
0
"As if to prove the"
ratio
0.01
100.00%
Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences
50
technicalSentenceCount
0
matches
(empty)
0.00%
Useless dialogue additions
Target: ≤5% dialogue tags with trailing filler fragments
totalTags
16
uselessAdditionCount
5
matches
0
"Silas said, a ghost of a smile on his lips"
1
"Silas asked, his gaze direct, piercing"
2
"he murmured, more to himself than to her"
3
"she said, the words barely a whisper"
4
"he said, his voice rough with emotion"
100.00%
Dialogue tag variety (said vs. fancy)
Target: ≤10% fancy dialogue tags
totalTags
9
fancyCount
1
fancyTags
0
"he murmured (murmur)"
dialogueSentences
29
tagDensity
0.31
leniency
0.621
rawRatio
0.111
effectiveRatio
0.069
75.2557%