NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-17-01KHMQEY7GTT0R7SKP5VHDM7PK
Run Details
Model:
Writer: Palmyra X5
Test:
Bad Writing Habits
Scenario:
Thriller: chase through city streets
Duration:
19.5s
Cost:
$0.010228
Provider:
Amazon Bedrock
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 London like it had a personal vendetta. Detective Harlow Quinn pressed forward against the downpour, her breath sharp in her lungs, boots splashing through flooded gutters. The city's neon bled into the wet pavement—red, gold, green—reflecting
fractured
signs and headlights that smeared into liquid light. She didn’t need to see clearly. She knew the outline of him: hooded, lean, moving with the desperate precision of someone who knew how to vanish. Tomás Herrera. She’d
watch
ed him slip out the back of *The Raven’s Nest* ten minutes ago, unnoticed by the bar’s regulars. She’d been observing the place for three weeks—quiet surveillance, no warrants, no backup. Just her gut and the ghost of Morris whispering in the silence between her thoughts. They’d found another body three days ago—pale, eyes blackened as if scorched from within, skin marked by symbols drawn in ash and old blood. No cause of death. No fingerprints. Just the faint scent of camphor and sulfur clinging to the air like a curse. And a partial print on the victim’s shirt: traced back to a discarded medical glove registered under the name of Tomás Herrera, NHS, terminated two years prior. She’d dug deeper. Found gaps—records with missing dates, a quiet ER report flagged *confidential*, a witness who claimed the man had pulled a woman out of cardiac arrest with “hands that glowed.” Nonsense. Of course. But Morris had said that once too—about a girl found floating upside down in the River Lea with her shadow crawling ahead of her. *You don’t know what you’re dealing with,* he’d said. And then he
was gone
. Quinn tightened the grip on her sidearm, holster half-open, heart pounding louder than the rain. She didn’t fire. Not yet. She needed answers, not another dead end wrapped in silence and paperwork. Herrera turned hard right into an alley between two shuttered shops. Quinn followed, gun now drawn, shoulders coiled. The alley dead-ended at a rusted fire escape and a graffiti-smeared wall. But he wasn’t there. She slowed, scanning. Her breath fogged in the cold. The rain drummed on dumpster lids, on concrete, on her collar. Then—a
flicker
. A shadow shifted behind a tattered tarpaulin hanging from a ventilation grate. She moved silently, training overriding exhaustion. The tarp came down, revealing a narrow set of iron steps descending into the earth. No sign. No railing. Just damp stone and the stench of mildew and something sharper—ozone, maybe, or burnt hair. Her wrist
watch
ticked—11:43 p.m. The green glow of her phone screen showed no signal. She hesitated. Orders said *no off-grid entries*. Not without clearance. Not after the Blackfriars incident. Not after Morris. But the steps led down. And Herrera was below. She took a breath. Pulled the bone token from her coat pocket—the one she’d lifted from the last victim’s clenched hand. It was smooth, pale,
etched
with a spiral. She didn’t know what it was. But the dead man hadn’t carried anything else. Quinn descended. The air changed fast—warmer, thicker, humming with something beneath hearing, like a wire str
etched
too tight. The steps fed into a tunnel, arched with cracked tile, lit by
flicker
ing sconces that burned with an odd blue flame. The walls
were lined
with symbols—some carved, some painted—none matching any language she recognized. Her
watch
stuttered, hands jerking forward, then stuttering back. She pressed on. The tunnel opened into a cavernous space beneath the city—a forgotten Tube station, its platforms cracked, tiles bulging. But it was alive now. Stalls str
etched
into the darkness, selling wares under dim lanterns: vials of smoking liquid, bundles of dried herbs tied with human hair, clocks with no hands, keys that melted when touched. Figures moved between them—some human, some not. A woman with eyes like polished onyx. A man whose shadow didn’t match his shape. A child who held a caged wind. *The Veil Market.* She’d heard the rumors. Ghost stories. Urban myth. Morris had believed it was real. She tightened her grip on the token. Blended in. Kept her head low, shoulders squared, eyes scanning. Uniformed authority wouldn’t help here. If anything, it’d get her killed. Then—movement to her left. Herrera stood at a stall exchanging a leather pouch for a small glass vial. He touched the Saint Christopher medal at his throat, muttered something under his breath. The vendor—a tall figure wrapped in moth-eaten
velvet
—nodded, then turned away like he’d never been there. Quinn moved. She stepped in front of him as he turned, gun low but visible. “Tomás Herrera,” she said, voice steady. “Metropolitan Police. You’re under arrest.” He froze. Not with fear—with resignation, like he’d known this moment
was coming
. “You shouldn’t be here, Detective,” he said, rain still clinging to his curls, his warm brown eyes searching hers. “This place… it doesn’t forgive trespassers.” “I’m not here to forgive,” she said. “I’m here for answers. Three bodies. All linked to you. All with marks no coroner can explain. What the hell are you involved in?” He glanced past her—toward the market, now eerily quiet. Vendors were closing stalls. Figures receding into tunnels. The air thickened, pressure building in her ears. “They’re
watch
ing,” he murmured. “And you’re holding a police badge in a place that eats badges for breakfast.” “I have the token,” she said, lifting it. “I got in, didn’t I?” “Luck,” he said. “And it’s run out.” A sound—a low chime, like glass struck in a tomb. From the far end of the platform, figures emerged. Hooded. Silent. Carrying rods of black iron. Their feet didn’t touch the ground. Herrera stepped forward, placing himself between her and them. “They don’t want you,” he said. “They want *her*.” “Who?” “The woman who died on my table last month. The one you’re carrying her token for. She wasn’t just a case. She was a guardian. And you’ve just walked in with her key.” Quinn’s pulse spiked. The token felt too warm in her palm. “I didn’t know,” she said. “I know,” he replied. “But they don’t care.” He turned to face the approaching figures. Reached into his coat. “I can get you out,” he said. “But you have to trust me.” Her finger hovered over the trigger. Eighteen years of procedure screamed at her—*arrest him, cuff him, bring him in.* Rules. Order. Control. But the man standing in front of her had just become a shield. She thought of Morris. His last words: *Some doors, Harlow, you don’t want to open. Some truths, you’re better not knowing.* But she’d come this far. And she was tired of not knowing. Quinn lowered her gun. “Tell me how,” she said. Herrera exhaled. Nodded. “Stay close. Don’t look them in the eyes. And whatever you do—don’t let go of that token.” Around them, the blue flames went out. Darkness swallowed the platform. And the chase began again—not through alleys, but through the veins of something older, hungrier, and buried deep beneath the city’s bones.
created_at
"2026-02-17T02:39:43.384137+00:00"
2/17/2026, 2:39:43 AM
model
"writer/palmyra-x5-20250428"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
417
moderation_latency
160
generation_time
19293
tokens_prompt
914
tokens_completion
1930
native_tokens_prompt
826
native_tokens_completion
1622
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
0.0102276
router
(null)
provider_responses
0
status
200
is_byok
false
latency
417
endpoint_id
"c684c4e7-1b77-4ca5-9516-7ecc66e2c455"
provider_name
"Amazon Bedrock"
model_permaslug
"writer/palmyra-x5-20250428"
api_type
"completions"
id
"gen-1771295963-nX8TKhpdL29gQzBsYaJm"
upstream_id
(null)
total_cost
0.0102276
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
11
adverbTagCount
1
adverbTags
0
"he said, rain still clinging to his curls, his warm brown eyes searching hers [still]"
100.0000%
AI-ism adverb frequency
Target: <1% AI-ism adverbs (57 tracked)
wordCount
1180
totalAiIsmAdverbs
0
found
(empty)
highlights
(empty)
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)
66.1017%
AI-ism word frequency
Target: <1% AI-ism words (140 tracked)
wordCount
1180
totalAiIsms
4
found
0
word
"flicker"
count
1
1
word
"velvet"
count
1
2
word
"fractured"
count
1
3
word
"etched"
count
1
highlights
0
"flicker"
1
"velvet"
2
"fractured"
3
"etched"
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
126
matches
(empty)
100.0000%
Filter word density
Target: ≤12% sentences with filter/hedge words
filterCount
1
hedgeCount
0
narrationSentences
126
filterMatches
0
"watch"
hedgeMatches
(empty)
100.0000%
Overuse of "that" (subordinate clause padding)
Target: ≤10% sentences with "that" clauses
thatCount
0
totalSentences
138
matches
(empty)
100.0000%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
62
mean
18.58
std
18.05
cv
0.971
sampleLengths
0
9
1
64
2
2
3
47
4
72
5
32
6
27
7
9
8
5
9
32
10
34
11
34
12
6
13
34
14
14
15
2
16
16
17
5
18
4
19
43
20
2
21
60
22
3
23
83
24
3
25
14
26
28
27
4
28
44
29
2
30
13
31
11
32
13
33
25
34
31
35
25
36
18
37
13
38
7
39
32
40
9
41
9
42
1
43
33
44
11
45
5
46
8
47
11
48
13
49
6
100.0000%
Passive voice overuse
Target: ≤5% passive sentences
passiveCount
2
totalSentences
126
matches
0
"was gone"
1
"were lined"
100.0000%
Past progressive (was/were + -ing) overuse
Target: ≤10% past progressive verbs
pastProgressiveCount
1
totalVerbs
190
matches
0
"was coming"
100.0000%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
1018
adjectiveStacks
0
stackExamples
(empty)
adverbCount
29
adverbRatio
0.02848722986247544
lyAdverbCount
3
lyAdverbRatio
0.0029469548133595285
100.0000%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
138
echoCount
0
echoWords
(empty)
100.0000%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
138
mean
8.35
std
6.74
cv
0.807
sampleLengths
0
9
1
19
2
19
3
6
4
20
5
2
6
19
7
13
8
15
9
24
10
4
11
2
12
15
13
27
14
3
15
29
16
1
17
2
18
24
19
9
20
5
21
15
22
3
23
2
24
12
25
11
26
7
27
12
28
4
29
3
30
6
31
11
32
2
33
12
34
6
35
15
36
2
37
2
38
15
39
4
40
10
41
2
42
5
43
3
44
5
45
3
46
5
47
4
48
4
49
17
56.2802%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
14
diversityRatio
0.39855072463768115
totalSentences
138
uniqueOpeners
55
84.0909%
Dialogue tag variety (said vs. fancy)
Target: ≤30% fancy dialogue tags
totalTags
11
fancyCount
4
fancyTags
0
"she said, voice steady (say, voice)"
1
"he said, rain still clinging to his curls, his warm brown eyes searching hers (say, still cling search)"
2
"he murmured (murmur)"
3
"she said, lifting it (say, lift)"
94.4984%