NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-17-01KHMQE6P3TZPB8Z25CK5MYPFD
Run Details
Model:
MoonshotAI: Kimi K2.5
Test:
Bad Writing Habits
Scenario:
Thriller: chase through city streets
Duration:
1m 20s
Cost:
$0.010632
Provider:
Together
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 hammered against the green neon sign above The Raven’s Nest, turning Soho’s cobblestones into a mirror of
fractured
light. Detective Harlow Quinn stood in the mouth of an alley opposite the bar, her spine pressed to wet brick,
watch
ing the door. Eighteen years on the force had taught her patience, but the leather
watch
on her left wrist felt heavier than usual, its worn strap drinking in the downpour. Three years to the day since Morris had died in circumstances the coroner had described as “unexplained cardiac event” and she’d described as bullshit. The file was still open on her desk, dog-eared, whispering. The door opened. Tomás Herrera stepped out, olive skin pale under the streetlamps, short curls plastered to his skull. Even from forty yards, Quinn saw the
glint
of the Saint Christopher medallion against his chest and the pale track of the scar snaking along his left forearm as he rolled down his sleeves. Former paramedic. Current criminal. He paused, scanning the street with the hyper-awareness of prey animals, and then he ran. Quinn was moving before her conscious mind caught up, her 5'9" frame cutting through the rain with military precision. Her boots struck the pavement in tight, controlled bursts, salt-and-pepper hair—cropped close to her skull for exactly this reason—shedding water in flicking droplets. She’d been
watch
ing the clique for months, and Herrera was the soft spot, the one who treated their wounded, who carried guilt in the slope of his shoulders. He was fast, though—twenty-nine years old, long legs, and he knew these streets like prayer. He darted left onto Wardour Street, weaving between late-night revelers who stumbled from curry houses. Quinn followed, her sharp jaw set, her breath steady despite the hammer in her chest. She was forty-one, not twenty-five, but she compensated with economy. No wasted motion. Her left arm pistoned, the leather
watch
band chafing against her wristbone as she pumped up speed. Herrera glanced back. Their eyes met—warm brown against her own dark irises—and then he vaulted a railing and dropped onto the access road running parallel to the Tottenham Court Road construction site. Quinn swore under her breath and followed, her palms slapping cold metal, the impact jarring up through her knees as she landed. Mud squelched. The smell of diesel and wet earth choked the air. He was heading north. The chase narrowed into the service alleys behind Camden High Street, where the streetlamps were broken teeth of sodium light. Quinn’s thighs burned. She could hear his breathing now, ragged, and the clink of the medallion bouncing against his sternum. He was making for the old Underground entrance—the one boarded up after the 1980s expansion, marked only by a
mosaic
sign half-covered in graffiti. He knew she wouldn’t follow him down there. That’s what the look back had been. A dare. Quinn reached the entrance as Herrera pried loose a plywood sheet and slid into the dark maw of the abandoned station. She stood at the threshold, rain drilling the pavement behind her, her chest heaving. The blackness inside smelled wrong—not like urine and dust, which would be natural for a derelict Tube station, but like ozone and burning sage, like copper pennies held under the tongue. Her hand went to her
watch
. Morris had died in a place like this. Not here exactly, but somewhere that smelled of this wrongness. She’d found him in a warehouse in Whitechapel, eyes wide, skin marked with patterns the forensic team couldn’t explain, his mouth open in a scream that had frozen solid in his throat. She’d never understood what he’d stumbled into. Only that it had killed him. Below, she heard
footsteps
scrambling down the escalator shafts—the old wooden ones, rotted through in places. She stepped inside. The darkness enveloped her like a fever. Her eyes adjusted
slowly
, picking out phosphorescent glows that had no business existing in a decommissioned station. The air grew warmer, humid, fungal. As she descended the fixed ladder—escalator steps long since removed—her fingers found graffiti that wasn’t graffiti at all, but gouges in the concrete, symbols that made her teeth ache. At the platform level, she crouched behind a crumbling pillar and looked out. The Veil Market. It shouldn’t have fit. The space yawned wider than the architecture allowed, archways receding into shadows that moved independently of the light sources—glowing orbs drifting between stalls made of scrap metal and
velvet
. Figures haggled in whispers over vials of swirling liquid, over bundles of herbs that smoked in the damp air, over information traded on slips of paper that
seemed to
burn as they were handed over. The bone token requirement was visible at the entrance arch: a skeletal hand carved from ivory or actual bone, where visitors pressed small, yellowed discs to gain passage. Herrera stood at a stall near the tracks, his scar visible as he pushed up his sleeve to accept something small and glass from a vendor whose face was hidden beneath a deep hood. He
looked scared
. Not of Quinn. Of what he was handling. Quinn’s hand drifted to her hip, to the service weapon holstered there. The
weight
was familiar, comforting, real. But the market wasn’t real. Not in the way her training had prepared her for. This was the world Morris had found. The supernatural substrate beneath London’s skin that she’d dismissed as trauma-induced hallucination in the witness reports she’d read, trying to solve his death. Her
watch
ticked, audible in the sudden stillness. A vendor turned, eyes
gleaming
with too many pupils, sensing her heat, her pulse, her solidity. Herrera looked up. Saw her. His expression wasn’t surprised. It was resigned. He held up the small vial—something amber, something that moved against gravity—and shook his head
slowly
. Warning her. Or begging her. Quinn stepped out from behind the pillar. Her boots crunched on debris that might have been rat bones or something older. The air pressure changed, pressing against her eardrums. She was uninvited, untokened, armed, and alone forty feet beneath Camden in a place that moved every full moon to avoid exactly this kind of intrusion. She thought of Morris’s body in the morgue, the incomprehensible burns on his fingertips. She walked toward the archway. Toward the bone token gate. Toward Herrera, and whatever truth had killed her partner. The decision was already made. It had been made three years ago, standing over a body that science couldn’t explain. Quinn reached into her jacket, withdrew her badge—not bone, but metal, stamped with the authority of the Metropolitan Police—and stepped into the market.
created_at
"2026-02-17T02:40:19.707542+00:00"
2/17/2026, 2:40:19 AM
model
"moonshotai/kimi-k2.5-0127"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
441
moderation_latency
(null)
generation_time
79498
tokens_prompt
914
tokens_completion
4354
native_tokens_prompt
825
native_tokens_completion
3650
native_tokens_completion_images
(null)
native_tokens_reasoning
2612
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
0.0106325
router
(null)
provider_responses
0
id
"oXgMguh-2dTqGa-9cf1e967153c9657"
status
200
is_byok
false
latency
441
endpoint_id
"3438c28e-ebcb-4f46-8bab-f133912d5297"
provider_name
"Together"
model_permaslug
"moonshotai/kimi-k2.5-0127"
api_type
"completions"
id
"gen-1771295939-fgSVOdyk24A08rBdATmf"
upstream_id
"oXgMguh-2dTqGa-9cf1e967153c9657"
total_cost
0.0106325
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Together"
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
0
adverbTagCount
0
adverbTags
(empty)
82.0949%
AI-ism adverb frequency
Target: <1% AI-ism adverbs (57 tracked)
wordCount
1117
totalAiIsmAdverbs
2
found
0
adverb
"slowly"
count
2
highlights
0
"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)
37.3321%
AI-ism word frequency
Target: <1% AI-ism words (140 tracked)
wordCount
1117
totalAiIsms
7
found
0
word
"mosaic"
count
1
1
word
"gleaming"
count
1
2
word
"glint"
count
1
3
word
"velvet"
count
1
4
word
"fractured"
count
1
5
word
"weight"
count
1
6
word
"footsteps"
count
1
highlights
0
"mosaic"
1
"gleaming"
2
"glint"
3
"velvet"
4
"fractured"
5
"weight"
6
"footsteps"
100.0000%
Cliche density
Target: ≤1 cliche(s) per 800-word window
totalCliches
0
maxInWindow
0
found
(empty)
highlights
(empty)
55.5556%
Emotion telling (show vs. tell)
Target: ≤3% sentences with emotion telling
emotionTells
1
narrationSentences
12
matches
0
"looked scared"
7.2464%
Filter word density
Target: ≤12% sentences with filter/hedge words
filterCount
3
hedgeCount
1
narrationSentences
12
filterMatches
0
"watch"
1
"watch. watch"
hedgeMatches
0
"seemed to"
100.0000%
Overuse of "that" (subordinate clause padding)
Target: ≤10% sentences with "that" clauses
thatCount
0
totalSentences
12
matches
(empty)
100.0000%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
27
mean
40.44
std
30.45
cv
0.753
sampleLengths
0
104
1
3
2
69
3
85
4
59
5
66
6
4
7
64
8
17
9
66
10
69
11
16
12
3
13
59
14
13
15
3
16
96
17
45
18
63
19
24
20
5
21
28
22
55
23
14
24
19
25
20
26
23
77.7778%
Passive voice overuse
Target: ≤5% passive sentences
passiveCount
1
totalSentences
12
matches
0
"were handed was hidden was resigned. been made"
100.0000%
Past progressive (was/were + -ing) overuse
Target: ≤10% past progressive verbs
pastProgressiveCount
3
totalVerbs
189
matches
0
"was heading was making was handling"
100.0000%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
1114
adjectiveStacks
0
stackExamples
(empty)
adverbCount
20
adverbRatio
0.017953321364452424
lyAdverbCount
8
lyAdverbRatio
0.00718132854578097
100.0000%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
12
echoCount
0
echoWords
(empty)
100.0000%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
12
mean
91
std
248.94
cv
2.736
sampleLengths
0
20
1
22
2
28
3
24
4
10
5
3
6
16
7
34
8
2
9
2
10
15
11
916
91.6667%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
1
diversityRatio
0.9166666666666666
totalSentences
12
uniqueOpeners
11
100.0000%
Dialogue tag variety (said vs. fancy)
Target: ≤30% fancy dialogue tags
totalTags
0
fancyCount
0
fancyTags
(empty)
85.3926%