NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY7WH86D4C1B7TRPEGCNP89
Run Details
Model:
Gemma 3 27B
Test:
Bad Writing Habits
Scenario:
Literary fiction: old friends reunite
Duration:
1m 20s
Cost:
$0.000376
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
22.22%
Adverbs in dialogue tags
(details)
78.01%
AI-ism adverb frequency
(details)
100.00%
AI-ism character names
(details)
100.00%
AI-ism location names
(details)
19.35%
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)
83.33%
Name drop frequency
(details)
85.90%
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)
83.78%
Passive voice overuse
(details)
100.00%
Past progressive (was/were + -ing) overuse
(details)
100.00%
Em-dash & semicolon overuse
(details)
99.98%
Purple prose (modifier overload)
(details)
100.00%
Repeated phrase echo
(details)
100.00%
Sentence length variance
(details)
70.80%
Sentence opener variety
(details)
70.92%
Adverb-first sentence starts
(details)
11.49%
Pronoun-first sentence starts
(details)
2.55%
Subject-first sentence starts
(details)
53.19%
Subordinate conjunction sentence starts
(details)
97.26%
Technical jargon density
(details)
11.36%
Useless dialogue additions
(details)
100.00%
Dialogue tag variety (said vs. fancy)
(details)
79.6721%
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 chipped Formica of the
bar was cool against Rory’s elbows.
Friday night at The Raven’s
Nest meant a slow burn, mostly regulars and the occasional lost tourist lured in by the green neon glow.
Silas, leaning against the back
bar polishing glasses with a practiced swirl, hadn’t said a word in the last ten minutes, a
comfortable
arrangement neither of them needed to
disrupt
.
It was a good night
for observing, for letting the city noise blur outside the heavy
velvet
curtains.
She’d clocked out an hour
ago, the smell of soy sauce and ginger still clinging to her jacket. Another hundred deliveries for Yu-Fei, another hundred smiles plastered on for impatient customers.
It wasn't glamorous, but it
paid the rent on the flat above the bar, and Silas didn’t ask questions.
Some debts were paid with
more than just currency.
The bell above the door
chimed, a discordant jangle that cut through the low hum of conversation.
Rory glanced up, automatically assessing
the newcomer. Tall. Broad shoulders.
The kind of build that
suggested a life lived outside, exposed to the elements.
Then he turned, and a
jolt ran through her that had nothing to do with caffeine or exhaustion. It was Ben. Ben Alistair.
The boy she’d known since
primary school, the boy whose hand she’d held during particularly bad thunderstorms, the boy she’d sworn she’d one day run away to Paris with. But this wasn’t the Ben she remembered.
The sandy blonde hair was
gone, replaced by a severe, close-cropped cut the color of slate.
His jaw, once soft, was
now
etched
with hard lines.
A network of fine scars
crisscrossed the back of his hand, barely visible beneath the cuff of his worn leather jacket. His eyes, once the bright, optimistic green of a spring meadow,
were shadowed
, guarded.
They scanned the room, lingering
on her for a fraction of a second before moving on,
as if trying to place a vaguely familiar face
.
He didn’t seem to recognize
her. Silas, however, had noticed.
He’d stopped polishing glasses and
was regarding
Ben with a quiet appraisal, his hazel
eyes narrowed
. The silver signet ring on his right hand caught the dim light. Ben settled onto a barstool a few feet down from Rory, the leather creaking in protest.
He signaled to Silas with
a curt nod. “Whiskey. Neat.” Silas moved with his usual unhurried grace, pouring the drink with steady hands. “Long time no see,”
he said, his voice a
low rumble. Ben’s eyes
flickered
to Silas, a
flicker
of something that might have been recognition, or perhaps just polite acknowledgment. “You know me, Silas?” Silas chuckled, a dry, rasping sound. “This is Soho, Mr. Alistair. A man can’t swing a cat without hitting someone’s past.”
He slid the glass across
the bar. Ben didn’t respond, simply took a long swallow of whiskey.
He stared into the amber
liquid, his expression
unreadable
. Rory, her heart hammering against her ribs, finally found her voice. “Ben?”
He turned slowly, his gaze
meeting hers. For a long moment, he just stared, his face a mask of confusion. Then, a slow dawning. “Rory?”
He said her name like
he
was testing
it, unsure if it still fit. “Bloody hell. Rory Carter. Is that
really
you?” “Last time I checked,”
she said, trying for a
lightness she didn’t feel.
He managed a ghost of
a smile, a
fleeting
expression that didn’t
quite reach his eyes
. “You haven’t changed a bit.” It was a lie. She had changed. The easy confidence of their youth had
been chipped
away, replaced by a cautious reserve. She touched her wrist unconsciously, her fingers
tracing
the faint crescent scar, a
reminder
of a childhood fall and, lately, of everything else she’d fallen from. “Neither have you,” she countered, knowing it was equally false. “Oh, I assure you,” he said, his voice laced with a weary irony, “I’ve changed plenty.” He took another drink, the whiskey burning its way down his throat. There was a loaded
silence
. Rory struggled to find something innocuous to say, something to bridge the chasm of years and
unspoken
history. “What brings you to Soho?” she asked, settling on the safest possible question. His gaze drifted to the maps that plastered the walls of The Raven’s Nest, ancient cartography mixed with modern city plans. “Business,” he said vaguely. "I'm...consulting." “Consulting,” she
echoed
, raising an
eyebrow
. Ben Alistair, the aspiring architect, now a consultant. It didn’t quite fit. “Sounds fancy.” He shrugged. “Pays the bills. What about you? Still sketching designs in the margins of your textbooks?” She laughed, a short, brittle sound. “Textbooks are a distant memory. I deliver Chinese food now.” His expression softened
slightly
. “Really? That’s… unexpected.” “Life has a way of being unexpected.” She paused, then ventured, “What happened? You just… vanished after uni. No calls, no letters. My mum was worried sick.” He flinched, a barely perceptible movement. “It’s a long story.” “We have time,” Rory said
softly
, recognizing the guardedness. Silas, ever observant, had quietly resumed polishing glasses, but she sensed his attention was still fixed on them. Ben hesitated, then sighed, running a hand through his cropped hair. “I joined up. Army. Thought I needed… a purpose.” “Army?” The word felt foreign on her tongue. Ben, the pacifist, the boy who wouldn’t hurt a fly. “It seemed like a good idea at the time.” He took another swill of whiskey. “Turns out, I was wrong.” “Where did you go?” He avoided her gaze. “Places. Doesn’t matter. I did things I’m not proud of.” “Things?” “Let’s just say it wasn’t building hospitals. I… I got involved in some operations. Covert stuff.” He finally looked at her, his eyes filled with a dark, haunted quality. “It changes you, Rory. You see things, you do things… you come back a different person.” She remembered his hands, skilled at delicate pencil work, capable of conjuring beauty from a blank page.
Now they bore the scars
of something else entirely. “I can see that.” “You wouldn't believe it if I told you half of it.” He shook his head. "It's better left unsaid." “Was it worth it?”
she asked, the question hanging in the air between them
. He stared at his glass, his knuckles white as he gripped it. "Worth it to whom?" The
silence
returned, heavier now, fraught with unasked questions and regrets. Rory thought of the plans they'd made, the dreams they'd shared, all dissolving into the mist of years and bad decisions. She thought of Evan, the reason she’d fled to London, and the lightness that had
been stolen
from her. Both of them had lost their way. “I always thought you’d build something beautiful,”
she said quietly, her voice almost a whisper
. He looked up, a shadow of pain crossing his face. “And you? I thought you’d be in Paris, painting masterpieces.” “Paris felt a long way away,”
she said, her gaze dropping to her hands
. “It always does, doesn’t it?" Ben finished, a hollow
echo
of our forgotten dreams. He downed the rest of his whiskey in one gulp and placed the empty glass on the bar. “I should go.” He stood, taller and broader than she remembered, a stranger in a
familiar
frame. “It was…” she started, searching for the right word. “Complicated,” he finished for her. "It'll always be complicated." He nodded curtly at Silas, then turned and walked towards the door, disappearing into the London night. Rory watched him go, a
profound
sense of
loss washing over her.
Not just for the boy she remembered, but for the life they might have had
. Silas placed a fresh glass in front of her, filling it with a neat measure of gin. "Sometimes," he said, his voice low, "the past is best left undisturbed." Rory didn’t respond. She raised the glass to her lips, the cold liquid burning a path down her throat. The Raven's Nest felt colder now, the shadows deeper. The scent of old maps and regret
hung heavy in the air
. Some debts, she realized, could never
truly
be repaid. And some ghosts
were destined
to haunt you, no matter how far you ran.
created_at
"2026-02-20T19:19:35.954Z"
2/20/2026, 7:19:35 PM
model
"google/gemma-3-27b-it"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
213
moderation_latency
(null)
generation_time
79640
tokens_prompt
892
tokens_completion
2179
native_tokens_prompt
850
native_tokens_completion
1925
native_tokens_completion_images
(null)
native_tokens_reasoning
0
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
0.000376
router
(null)
provider_responses
0
endpoint_id
"8f22002c-c045-446f-a1b9-9896133536b8"
id
"chatcmpl-RUrF8XzZam5jRmlFfeIVHJW9"
is_byok
false
latency
58
model_permaslug
"google/gemma-3-27b-it"
provider_name
"DeepInfra"
status
200
api_type
"completions"
id
"gen-1771615175-LOQwF9XpyVMqrtAQUH93"
upstream_id
"chatcmpl-RUrF8XzZam5jRmlFfeIVHJW9"
total_cost
0.000376
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
22.22%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
22
adverbTagCount
4
adverbTags
0
"he said vaguely [vaguely]"
1
"She paused then [then]"
2
"Rory said softly [softly]"
3
"she said quietly [quietly]"
dialogueSentences
45
tagDensity
0.489
leniency
0.978
rawRatio
0.182
effectiveRatio
0.178
78.01%
AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount
1364
totalAiIsmAdverbs
6
found
0
adverb
"slowly"
count
1
1
adverb
"really"
count
2
2
adverb
"slightly"
count
1
3
adverb
"softly"
count
1
4
adverb
"truly"
count
1
highlights
0
"slowly"
1
"really"
2
"slightly"
3
"softly"
4
"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)
19.35%
AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount
1364
totalAiIsms
22
found
0
word
"comfortable"
count
1
1
word
"disrupt"
count
1
2
word
"velvet"
count
1
3
word
"etched"
count
1
4
word
"scanned"
count
1
5
word
"familiar"
count
2
6
word
"flickered"
count
1
7
word
"flicker"
count
1
8
word
"unreadable"
count
1
9
word
"fleeting"
count
1
10
word
"tracing"
count
1
11
word
"reminder"
count
1
12
word
"silence"
count
2
13
word
"unspoken"
count
1
14
word
"echoed"
count
1
15
word
"eyebrow"
count
1
16
word
"whisper"
count
1
17
word
"echo"
count
1
18
word
"profound"
count
1
19
word
"sense of"
count
1
highlights
0
"comfortable"
1
"disrupt"
2
"velvet"
3
"etched"
4
"scanned"
5
"familiar"
6
"flickered"
7
"flicker"
8
"unreadable"
9
"fleeting"
10
"tracing"
11
"reminder"
12
"silence"
13
"unspoken"
14
"echoed"
15
"eyebrow"
16
"whisper"
17
"echo"
18
"profound"
19
"sense of"
100.00%
Cliché density
Target: ≤1 cliche(s) per 800-word window
totalCliches
2
maxInWindow
1
found
0
label
"eyes widened/narrowed"
count
1
1
label
"hung in the air"
count
1
highlights
0
"eyes narrowed"
1
"hung heavy in the air"
100.00%
Emotion telling (show vs. tell)
Target: ≤3% sentences with emotion telling
emotionTells
0
narrationSentences
98
matches
(empty)
100.00%
Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount
0
hedgeCount
1
narrationSentences
98
filterMatches
(empty)
hedgeMatches
0
"seem to"
100.00%
Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences
121
gibberishSentences
0
adjustedGibberishSentences
0
longSentenceCount
0
runOnParagraphCount
0
giantParagraphCount
0
wordSaladCount
0
repetitionLoopCount
0
controlTokenCount
0
maxSentenceWordsSeen
30
ratio
0
matches
(empty)
100.00%
Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans
0
markdownWords
0
totalWords
1362
ratio
0
matches
(empty)
100.00%
Missing dialogue indicators (quotation marks)
Target: ≤10% speech attributions without quotation marks
totalAttributions
18
unquotedAttributions
0
matches
(empty)
83.33%
Name drop frequency
Target: ≤1.0 per-name mentions per 100 words
totalMentions
44
wordCount
1097
uniqueNames
11
maxNameDensity
1
worstName
"Ben"
maxWindowNameDensity
2.5
worstWindowName
"Silas"
discoveredNames
Formica
1
Rory
9
Raven
3
Nest
3
Yu-Fei
1
Silas
10
Ben
11
Alistair
2
Paris
1
Evan
1
London
2
persons
0
"Rory"
1
"Raven"
2
"Nest"
3
"Silas"
4
"Ben"
5
"Alistair"
6
"Evan"
places
0
"Paris"
1
"London"
globalScore
0.999
windowScore
0.833
85.90%
Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences
78
glossingSentenceCount
2
matches
0
"as if trying to place a vaguely familiar face"
1
"quite reach his eyes"
100.00%
"Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches
1
per1kWords
0.734
wordCount
1362
matches
0
"Not just for the boy she remembered, but for the life they might have had"
100.00%
Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount
0
totalSentences
121
matches
(empty)
100.00%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
55
mean
24.76
std
18.67
cv
0.754
sampleLengths
0
83
1
59
2
59
3
3
4
31
5
93
6
6
7
32
8
26
9
24
10
23
11
28
12
19
13
12
14
24
15
23
16
13
17
21
18
48
19
10
20
28
21
23
22
13
23
26
24
20
25
17
26
16
27
7
28
27
29
10
30
27
31
20
32
18
33
20
34
4
35
14
36
1
37
45
38
26
39
4
40
19
41
14
42
16
43
58
44
15
45
20
46
14
47
35
48
14
49
9
83.78%
Passive voice overuse
Target: ≤2% passive sentences
passiveCount
6
totalSentences
98
matches
0
"were paid"
1
"was gone"
2
"were shadowed"
3
"been chipped"
4
"been stolen"
5
"were destined"
100.00%
Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount
2
totalVerbs
181
matches
0
"was regarding"
1
"was testing"
100.00%
Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount
0
semicolonCount
0
flaggedSentences
0
totalSentences
121
ratio
0
matches
(empty)
99.98%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
1099
adjectiveStacks
0
stackExamples
(empty)
adverbCount
43
adverbRatio
0.039126478616924476
lyAdverbCount
22
lyAdverbRatio
0.020018198362147407
100.00%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
121
echoCount
0
echoWords
(empty)
100.00%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
121
mean
11.26
std
6.9
cv
0.613
sampleLengths
0
11
1
24
2
30
3
18
4
18
5
13
6
19
7
9
8
17
9
7
10
1
11
2
12
14
13
18
14
3
15
2
16
29
17
7
18
16
19
10
20
21
21
14
22
25
23
6
24
4
25
16
26
12
27
16
28
8
29
2
30
13
31
11
32
19
33
4
34
6
35
22
36
10
37
9
38
11
39
1
40
7
41
13
42
4
43
15
44
8
45
13
46
16
47
5
48
4
49
3
70.80%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
1
diversityRatio
0.4297520661157025
totalSentences
121
uniqueOpeners
52
70.92%
Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount
2
totalSentences
94
matches
0
"Then he turned, and a"
1
"Then, a slow dawning."
ratio
0.021
11.49%
Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount
49
totalSentences
94
matches
0
"It was a good night"
1
"She’d clocked out an hour"
2
"It wasn't glamorous, but it"
3
"It was Ben."
4
"His jaw, once soft, was"
5
"His eyes, once the bright,"
6
"They scanned the room, lingering"
7
"He didn’t seem to recognize"
8
"He’d stopped polishing glasses and"
9
"He signaled to Silas with"
10
"he said, his voice a"
11
"He slid the glass across"
12
"He stared into the amber"
13
"He turned slowly, his gaze"
14
"He said her name like"
15
"she said, trying for a"
16
"He managed a ghost of"
17
"It was a lie."
18
"She had changed."
19
"She touched her wrist unconsciously,"
ratio
0.521
2.55%
Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount
86
totalSentences
94
matches
0
"The chipped Formica of the"
1
"Friday night at The Raven’s"
2
"Silas, leaning against the back"
3
"It was a good night"
4
"She’d clocked out an hour"
5
"Another hundred deliveries for Yu-Fei,"
6
"It wasn't glamorous, but it"
7
"Some debts were paid with"
8
"The bell above the door"
9
"Rory glanced up, automatically assessing"
10
"The kind of build that"
11
"It was Ben."
12
"The boy she’d known since"
13
"The sandy blonde hair was"
14
"His jaw, once soft, was"
15
"A network of fine scars"
16
"His eyes, once the bright,"
17
"They scanned the room, lingering"
18
"He didn’t seem to recognize"
19
"Silas, however, had noticed."
ratio
0.915
53.19%
Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount
1
totalSentences
94
matches
0
"Now they bore the scars"
ratio
0.011
97.26%
Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences
47
technicalSentenceCount
3
matches
0
"They scanned the room, lingering on her for a fraction of a second before moving on, as if trying to place a vaguely familiar face."
1
"Ben’s eyes flickered to Silas, a flicker of something that might have been recognition, or perhaps just polite acknowledgment."
2
"His gaze drifted to the maps that plastered the walls of The Raven’s Nest, ancient cartography mixed with modern city plans."
11.36%
Useless dialogue additions
Target: ≤5% dialogue tags with trailing filler fragments
totalTags
22
uselessAdditionCount
5
matches
0
"he said, his voice a low rumble"
1
"she asked, the question hanging in the air between them"
2
"she said quietly, her voice almost a whisper"
3
"she said, her gaze dropping to her hands"
4
"he said, his voice low,"
100.00%
Dialogue tag variety (said vs. fancy)
Target: ≤10% fancy dialogue tags
totalTags
11
fancyCount
0
fancyTags
(empty)
dialogueSentences
45
tagDensity
0.244
leniency
0.489
rawRatio
0
effectiveRatio
0
79.6721%