NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-17-01KHMQVVD7WTFGSM2FBRTSD6MX
Run Details
Model:
GPT-4.1 Nano
Test:
Bad Writing Habits
Scenario:
Thriller: chase through city streets
Duration:
9.3s
Cost:
$0.000600
Provider:
OpenAI
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 heavy sheets, pounding the cracked pavement with relentless fury. Harlow Quinn’s boots splashed through puddles large enough to swallow a man, each step
echo
ing with purpose. Her coat clung to her, soaked through to the skin beneath, but she hardly noticed. Her gaze stayed fixed on the figure darting ahead—a shadow slipping through the gloom with practiced ease. The suspect crouched beneath a battered awning, breath visible in the cold night air. He paused only long enough to cast a quick glance over his shoulder. Quinn caught the
flicker
of panic in his eyes, an animal caught in a trap. She closed the gap, her grip tightening instinctively on her belt as she surged forward. “Hold on!” she demanded, voice cutting through the rain, every word firm and direct. But he didn’t stop. Instead, he ducked lower, slipping between two dumpsters with a desperation that spoke of secrets too dangerous to reveal. Quinn
was prepared
for this level of chaos—the years
etched
into her bones. She slowed just a fraction, eyes narrowing as she timed her move. The suspect’s foot caught an uneven patch, a misstep that sent him sprawling onto the slick cobblestones. She was on him in an instant, grabbing his arm before he could scramble to his feet again. “Got you,” she said
tightly
, her eyes searching his face. “You’re making a mistake, you know that?” His face
was shadowed
, but she could see the flash of sweat and dirt smudged across his cheeks. His lips parted, voice tight with fury, and then he finally exhaled. “You don’t get it,” he whispered, voice hoarse. “They’re watching. Every move you make.” She tightened her grip, feeling the tremor in his muscles. “Who’s watching? What are you running from?” He hesitated, eyes darting to the dark entrance of an alley that plunged downward. A faint smell of damp concrete and something sharper—urine, or maybe something more
sinister
—drifted upward. Quinn’s instincts told her to stay put, to grab him and haul him back to the surface where the world was familiar. But her gut twisted,
a flicker of doubt
, an
echo
of her lost partner’s voice whispering warnings she’d yet to understand. The alley beckoned with an almost
magnetic
pull, as if it wanted her to step into its shadows. “Don’t,” she warned. Her finger pressed against his collarbone. “Come with me. You’re not getting away this time.” He looked at her with desperation, then
slowly
nodded. Their
footsteps
splashed into the darkness as they descended into the tunnel, the rain muffled behind the sealed mouth of the alley. The passage was narrow, lined with damp bricks and graffiti that
shimmered
under
flicker
ing bulbs. The air grew heavier,
tinged
with a metallic tang. Quinn kept her eyes sharp, every sense alive. She’d seen enough of London’s underbelly to know this wasn’t just a route for criminals; it could be something far worse. The tunnel twisted and descended further, slick with moss and slime. At the end, a rickety metal staircase spiraled down into complete darkness. The suspect hesitated, and Quinn didn’t wait. She pulled her flashlight from her belt, the beam cutting through the gloom. The staircase groaned beneath their
weight
, each step
echo
ing in the silence like a gunshot. Their breath
echo
ed louder. The faint hum of distant voices, or maybe machinery, grew clearer. Their feet hit a dirt floor, uneven and muddy, and the
flicker
ing light revealed a sprawling underground space. It was a market, or at least, what looked like one. Stalls built of battered wood and tarps cluttered the space. Amidst the shuffling bodies and low murmur of voices, Quinn spotted lanterns hung from rusted pipes, casting uneven pools of amber light. Shelves lined with bizarre objects—glowing vials, odd talismans, faded photographs—claimed every available surface. She stiffened. This was no ordinary black market. She’d heard whispers of the Veil Market—an illicit hub for supernatural dealings, hidden beneath the city. A place you came to buy banned alchemical substances, enchanted artifacts, or answers that might murder you on the spot. The suspect, now visibly tense,
started to
back away. Quinn’s grip on his sleeve tightened, her eyes scanning the entire space. Her military instincts screamed at her to stay cautious, to keep her hand near her weapon. But something about this scene—about what lurked just beyond the
flicker
ing edges of her vision—felt wrong. Like an
unspoken
weight
pressing down on her chest. Her mind leapt back to the case of her partner, Morris, lost under mysterious supernatural circumstances. Every step she’d taken to solve that had brought her closer to places like this, into shadows stitched thick with secrets she wasn’t ready to face. A figure stepped into view, cloaked and hooded, voice low and gravelly. “You’ve found us, detective.” Quinn drew her gun with a smooth, practiced motion, aiming it at the stranger. Her finger pressed the trigger guard, but she hesitated. She’d chased enough ghosts tonight to understand: sometimes, the ghost you feared was only the beginning. The suspect shrank behind her as the shadows thickened. Quinn’s eyes
flicked
across the market—over shifty eyes behind stalls, a crumpled cloth that
seemed to
shimmer momentarily, as if alive with it’s own magic. She knew somewhere here, hidden in this tangled underworld, was something darker than mere lawless traders. She saw him, finally, lurking at the edge—an
imposing
figure, cloaked in black, face obscured. Their eyes met across the chaos-littered space, and Harlow Quinn felt her pulse spike. This was beyond petty crime; this was something old, something lethal. Her instinct was to retreat, to force her way back up the staircase, to leave the shadows behind. But she was here now, and she knew—she’d crossed into something she wasn’t ready for. Not yet. The rain still hammered on above. Down here, beneath the city’s surface, the shadows waited—but this time, she was ready to face them.
created_at
"2026-02-17T02:46:36.618323+00:00"
2/17/2026, 2:46:36 AM
model
"openai/gpt-4.1-nano-2025-04-14"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
399
moderation_latency
(null)
generation_time
9114
tokens_prompt
914
tokens_completion
1575
native_tokens_prompt
811
native_tokens_completion
1297
native_tokens_completion_images
0
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
"completed"
usage
0.0005999
router
(null)
provider_responses
0
id
"resp_02f7b5fe3e67b8a0016993d6829038819590b5e77b9e02d440"
status
200
is_byok
false
latency
141
endpoint_id
"9251cee5-5503-4be9-9439-7ae21ff062a3"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-4.1-nano-2025-04-14"
api_type
"completions"
id
"gen-1771296386-Z5xD60Ot7WHt7V5SvfXa"
upstream_id
"resp_02f7b5fe3e67b8a0016993d6829038819590b5e77b9e02d440"
total_cost
0.0005999
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
50.0000%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
4
adverbTagCount
1
adverbTags
0
"she said tightly, her eyes searching his face [tightly]"
79.9599%
AI-ism adverb frequency
Target: <1% AI-ism adverbs (57 tracked)
wordCount
998
totalAiIsmAdverbs
2
found
0
adverb
"slowly"
count
1
1
adverb
"tightly"
count
1
highlights
0
"slowly"
1
"tightly"
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)
0.0000%
AI-ism word frequency
Target: <1% AI-ism words (140 tracked)
wordCount
998
totalAiIsms
18
found
0
word
"tinged"
count
1
1
word
"flicker"
count
2
2
word
"flicked"
count
1
3
word
"shimmered"
count
1
4
word
"sinister"
count
1
5
word
"imposing"
count
1
6
word
"unspoken"
count
1
7
word
"magnetic"
count
1
8
word
"weight"
count
2
9
word
"footsteps"
count
1
10
word
"amidst"
count
1
11
word
"echo"
count
1
12
word
"echoed"
count
1
13
word
"echoing"
count
2
14
word
"etched"
count
1
highlights
0
"tinged"
1
"flicker"
2
"flicked"
3
"shimmered"
4
"sinister"
5
"imposing"
6
"unspoken"
7
"magnetic"
8
"weight"
9
"footsteps"
10
"amidst"
11
"echo"
12
"echoed"
13
"echoing"
14
"etched"
100.0000%
Cliche density
Target: ≤1 cliche(s) per 800-word window
totalCliches
1
maxInWindow
1
found
0
label
"flicker of emotion"
count
1
highlights
0
"a flicker of doubt"
100.0000%
Emotion telling (show vs. tell)
Target: ≤3% sentences with emotion telling
emotionTells
0
narrationSentences
71
matches
(empty)
100.0000%
Filter word density
Target: ≤12% sentences with filter/hedge words
filterCount
0
hedgeCount
2
narrationSentences
71
filterMatches
(empty)
hedgeMatches
0
"started to"
1
"seemed to"
100.0000%
Overuse of "that" (subordinate clause padding)
Target: ≤10% sentences with "that" clauses
thatCount
0
totalSentences
76
matches
(empty)
63.6096%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
24
mean
41
std
15.28
cv
0.373
sampleLengths
0
62
1
57
2
37
3
60
4
17
5
44
6
17
7
29
8
61
9
18
10
31
11
53
12
43
13
48
14
56
15
44
16
37
17
67
18
16
19
39
20
50
21
40
22
35
23
23
100.0000%
Passive voice overuse
Target: ≤5% passive sentences
passiveCount
2
totalSentences
71
matches
0
"was prepared"
1
"was shadowed"
100.0000%
Past progressive (was/were + -ing) overuse
Target: ≤10% past progressive verbs
pastProgressiveCount
0
totalVerbs
182
matches
(empty)
94.0000%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
956
adjectiveStacks
1
stackExamples
0
"under mysterious supernatural circumstances."
adverbCount
35
adverbRatio
0.036610878661087864
lyAdverbCount
12
lyAdverbRatio
0.012552301255230125
100.0000%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
76
echoCount
0
echoWords
(empty)
100.0000%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
76
mean
12.95
std
5.32
cv
0.411
sampleLengths
0
13
1
17
2
15
3
17
4
14
5
13
6
15
7
15
8
14
9
4
10
19
11
13
12
12
13
17
14
18
15
10
16
7
17
18
18
12
19
8
20
6
21
10
22
7
23
14
24
15
25
22
26
21
27
18
28
3
29
6
30
9
31
9
32
22
33
15
34
9
35
8
36
21
37
11
38
12
39
7
40
13
41
15
42
4
43
11
44
18
45
11
46
10
47
22
48
13
49
2
56.1404%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
4
diversityRatio
0.3684210526315789
totalSentences
76
uniqueOpeners
28
0.0000%
Dialogue tag variety (said vs. fancy)
Target: ≤30% fancy dialogue tags
totalTags
4
fancyCount
4
fancyTags
0
"demanded (demand)"
1
"she said tightly, her eyes searching his face (say tightly, search)"
2
"he whispered, voice hoarse (whisper)"
3
"she warned (warn)"
79.0418%