NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-20-01M5C6HMJ2W93X1MDMVRR5KBKT
Run Details
Model:
DeepSeek V4.1 Flash 1017 (Reasoning, High)
Test:
Bad Writing Habits
Scenario:
Mystery: examining a crime scene
Duration:
34.5s
Cost:
$0.001263
Provider:
Wafer
Evaluation summary
Score
Evaluator
Details Link
100.00%
Adverbs in dialogue tags
(details)
86.13%
AI-ism adverb frequency
(details)
100.00%
AI-ism character names
(details)
100.00%
AI-ism location names
(details)
54.92%
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)
42.54%
Name drop frequency
(details)
100.00%
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)
86.99%
Passive voice overuse
(details)
58.16%
Past progressive (was/were + -ing) overuse
(details)
100.00%
Em-dash & semicolon overuse
(details)
99.69%
Purple prose (modifier overload)
(details)
100.00%
Repeated phrase echo
(details)
100.00%
Sentence length variance
(details)
52.13%
Sentence opener variety
(details)
0.00%
Adverb-first sentence starts
(details)
49.89%
Pronoun-first sentence starts
(details)
5.98%
Subject-first sentence starts
(details)
0.00%
Subordinate conjunction sentence starts
(details)
100.00%
Technical jargon density
(details)
45.00%
Useless dialogue additions
(details)
100.00%
Dialogue tag variety (said vs. fancy)
(details)
79.3808%
Overall
Model input
System
You are an expert fiction writer. Always keep the following rules in mind: - Write in past tense and use General English spelling, grammar, and colloquialisms/slang. - Write in active voice - Always follow the "show, don't tell" principle. - Avoid adverbs and cliches and overused/commonly used phrases. Aim for fresh and original descriptions. - Convey events and story through dialogue. - Mix short, punchy sentences with long, descriptive ones. Drop fill words to add variety. - Skip "he/she said said" dialogue tags and convey people's actions or face expressions through their speech - Avoid mushy dialog and descriptions, have dialogue always continue the action, never stall or add unnecessary fluff. Vary the descriptions to not repeat yourself. - Put dialogue on its own paragraph to separate scene and action. - Reduce indicators of uncertainty like "trying" or "maybe" When writing text: - NEVER conclude the scene on your own, follow the beat instructions very closely. - NEVER end with foreshadowing. - NEVER write further than what I prompt you with. - AVOID imagining possible endings, NEVER deviate from the instructions. - STOP EARLY if the continuation contains what was required in the instructions. You do not need to fill out the full amount of words possible.
User/Human
Take into account the following glossary of characters/locations/items/lore... when writing your response: <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="4414" name="Eva Kowalski" aliases="Eva, Evie" pronouns="she/her" species="Human" storyRole="Best Friend"> <eyeColor>green</eyeColor> <hair>curly red</hair> <height>5'4"</height> <age>26</age> <glasses>round glasses</glasses> <satchel>worn leather satchel full of books</satchel> <freckles>freckled complexion</freckles> <nervousHabit>tucking hair behind her left ear</nervousHabit> <description>Aurora's childhood friend and an occult researcher.</description> <backstory>Moved to London two years before Aurora and works as a research assistant at the British Museum's restricted archives. She holds a master's degree in Ancient History from Oxford.</backstory> </character> <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> <item id="2004" name="Veil Compass" aliases="the Compass, shadow compass" type="Tool" origin="The Veil Market"> <casingMaterial>small brass compass</casingMaterial> <faceMarkings>face etched with protective sigils</faceMarkings> <needle>points toward the nearest supernatural rift or portal</needle> <patina>casing has a patina of verdigris</patina> <creator>crafted by a Shade artisan</creator> <description>A compass attuned to supernatural energy.</description> </item> </codex> Continue the story using the following instructions: <instructions> Write a fully fleshed out scene of at least 1000 words in the mystery genre. A detective arrives at a crime scene in an unusual location. Something is wrong — the evidence doesn't add up. A colleague is present, offering a different interpretation. Through careful observation and deduction, the detective begins to see what others have missed. 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. </instructions>
Model response
Messages
Meta
AI
The smell hit first.
Not blood or decay, but a thick, greasy scent like a spent match
and wet earth, clinging to the air of the abandoned Tube station.
Water dripped somewhere in the
dark, a steady,
echoing
plink into a shallow pool.
Harlow Quinn ducked under the
police tape, her torch beam cutting a
stark
white line through the cavernous
gloom
of the platform. "Detective,"
a constable called, his voice
bouncing off the grimy tiles. She gave a curt nod, her eyes already sweeping the scene.
The body lay crumpled on
the cracked concrete, a splash of colour against the grey.
A man, mid-thirties, dressed in
a bespoke suit that cost more than Harlow's first car.
His face was a mask
of surprise, eyes wide, mouth slack. No visible wounds. No signs of a struggle.
A single item rested on
his chest: a small brass compass, its casing green with age. "Who found him?"
Harlow asked, crouching
. Her knees popped. She ignored it. "Cleaner,"
the constable said, jerking a
thumb towards the stairs. "Does the rounds once a week. Nearly had a heart attack."
Harlow pulled on a latex
glove and picked up the compass.
The face was etched with
strange symbols, not any maker's mark she recognised.
The needle spun lazily, not
pointing north.
She flicked her wrist, but
it just wobbled, indifferent.
A set of heavy footsteps
echoed
from the tunnel. "Quinn. They said you'd be down here." She didn't look up. "Sergeant."
DS Miller, a big man
with a perpetually worried expression, came to a halt beside her.
He smelled of cheap coffee
and damp wool. "Pathologist reckons time of death between midnight and two. No ID on him. But look at this."
He gestured to the body
. "No wallet, no phone. But he's got a watch on. Rolex. Worth a few grand, easy. Robbery gone wrong? But they left the watch. And they left that."
He nudged his chin at
the compass in her hand. "Amateurs panic," Harlow murmured, turning the compass over. The back was smooth, unadorned. "Or it wasn't a robbery." "Maybe they were interrupted," Miller offered. "The cleaner, perhaps. Or they just wanted something specific. Something in his pockets." Harlow stood, her gaze drifting past the body to the tunnel mouth. Faint, chalky symbols
were scrawled
on the curved brickwork, almost invisible in the shadows.
She walked towards them, her
torchlight catching the crude, looping lines.
They looked like the same
style as the etchings on the compass. "What do you make of the location, sir?" Miller asked, following her. "An abandoned station. Underground. Makes sense for a dodgy deal, doesn't it? Somewhere quiet." "Quiet," Harlow repeated.
Her eyes traced the line
of the platform. "This station closed sixty years ago. It's damp, it's mouldy, and it's not on any current map. How did a man in a two-thousand-pound suit find his way down here?" "Business, I suppose." "Business,"
she said again
. "Look at the dust."
She pointed her torch at
the floor. "Thick, undisturbed. Apart from our own footprints and the cleaner's, there's one clear set of prints leading to the body. But not away." Miller squinted. "What? So he came down here, and… what? The killer flew out?" "Or the killer was never here," Harlow said, turning back to the body.
She knelt again, her eyes
fixed on the man's expensive shoes. The leather was
perfectly
polished, save for a thin, even layer of dust on the soles. "He walked down here. That's it. He wasn't dragged." "So, he meets someone. They argue. They push him. He falls, hits his head." Miller gestured towards a faint bruise on the man's temple. "That’s a plausible head injury. A simple assault that went too far." "And the compass?" Miller shrugged. "He was holding it. It fell on him." Harlow shook her head, a slow, deliberate motion. "He's holding it in a place of honour. In a way that says, 'this is important.' A dying man doesn't place a trinket on his chest like a funeral offering. Someone else did that. After."
She stood up and took
a slow, careful walk around the perimeter of the scene, her eyes on the ground. The torch beam swept left to right, back and forth.
She stopped, ten feet from
the body. "Miller. Here." He lumbered over. "What?" She pointed at the concrete. "What do you see?" He stared. "Dust. Some grit. A few water droplets." "Look closer. Drag marks."
She traced a faint, smeared
path with her finger, barely visible against the grime. "Something was moved." She looked back toward the body, then followed the faint marks to a rusted metal casing bolted to the wall. An old signalling box. The door was
slightly
ajar. "Someone was here,"
Harlow said, more to herself than to Miller
. "Not just the victim. Someone came, did what they did, and left. They were careful. But not careful enough." She returned to the body and crouched beside it once more, her focus sharpening. The man's hands. The left one
was clutched
tight, knuckles white. She
gently
uncurled the fingers. In his palm was a small, sharp piece of black stone, shaped like a shard of glass. "What's that?" Miller asked, peering over her shoulder. "Doesn't look like it belongs down here." "It doesn't," Harlow said. She sniffed it. Nothing. She pressed the stone shard into an evidence bag. "Look at his jacket." The suit jacket was barely buttoned. The man was a fastidious dresser. He'd know how to wear a suit. But he'd put it back on in a hurry. The left pocket was half-tucked in. "He came here with something," Harlow said, thinking aloud. "Something he intended to trade. The compass, perhaps. But he didn't get the chance." "You think it was a drop gone wrong?" Miller asked. "He brought the compass, someone else brought whatever's in the stone. The deal went south. They fought. The other bloke did him in, took his prize, and left." "It fits," Harlow admitted. "But what was the prize? And why leave the compass? If you've just killed someone for their goods, you wouldn't leave a valuable object on their corpse. You'd take everything. Unless the compass wasn't what they were after. Unless it was…" She trailed off, her gaze falling on the compass once more. She picked up the evidence bag, holding it up to her torch. The needle
was still dancing
, a frantic, pointless pirouette. "Miller. You said he was found by the cleaner. What time?" "Just after six this morning." "And the cleaner said he "nearly had a heart attack." What does that mean? Was he scared by the body, or something else?" Miller frowned, thinking. "I didn't ask. He was pretty shaken." Harlow's
eyes narrowed
. "The symbols on the wall. Did anyone ask about those?" "What, the graffiti? Kids messing about, probably. They get everywhere." "Look at them again. Now look at the marks on the compass. It's not teens, Miller. This is deliberate. And it's not just decoration. This is a map. Or a key." She walked back to the wall, her finger
tracing
the chalky lines. It was a circle, bisected by a line, with a series of smaller symbols along the edge. It was crude, but it was a compass rose. The same symbols as the brass compass. And one of the symbols
was drawn
twice, with a deliberate, thick line. She hurried back to the body, the realisation hitting her like a physical blow. The compass needle
wasn't spinning
aimlessly. It
was jerking
, pointing towards the tunnel, then back towards the body, then up towards the ceiling. It
was agitated
. "He wasn't making a deal,"
Harlow said, her voice low, intense
. "He was a messenger. This isn't a murder. It's a message." "A message?" Miller's brow
furrowed
. "To who?" Harlow held up the compass. "This points to something. Something near." She looked at the chalk diagram, at the doubled symbol. She turned to the signalling box. "The compass is a key. He was supposed to leave it here for someone to find. But he was intercepted. He tried to hide the most important part… the stone shard… in his own hand. Because the compass itself was just the bait." "Bait for what?" "The same thing the killer wanted,"
Harlow said, her eyes scanning the dark
. "And if they haven't found it, they'll be back." She turned to Miller. "I need you to get forensics down here. Not a minute later. And check the tunnels. I need to know where they lead. The compass is pointing us to something, Miller. And whatever it is, it's still here."
created_at
"2026-10-20T08:23:47.019Z"
10/20/2026, 8:23:47 AM
model
"deepseek/deepseek-v4.1-flash-20261017"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
619
moderation_latency
(null)
generation_time
34513
tokens_prompt
1112
tokens_completion
2362
native_tokens_prompt
1059
native_tokens_completion
2188
native_tokens_completion_images
(null)
native_tokens_reasoning
116
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
service_tier
(null)
usage
0.00126344
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1792484627-poBmhn8L9yHtxRDVBW4O"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1792484627-h6oGsZGhNT19falmVlWJ"
upstream_id
"chatcmpl-f18e4811e47045d5949bcfeb96967ee3"
provider_responses
0
endpoint_id
"66d4851d-7738-4948-9d40-2ab6dc54f895"
id
"chatcmpl-f18e4811e47045d5949bcfeb96967ee3"
is_byok
false
latency
619
model_permaslug
"deepseek/deepseek-v4.1-flash-20261017"
provider_name
"Wafer"
status
200
total_cost
0.00126344
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Wafer"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
25
adverbTagCount
1
adverbTags
0
"she said again [again]"
dialogueSentences
63
tagDensity
0.397
leniency
0.794
rawRatio
0.04
effectiveRatio
0.032
86.13%
AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount
1442
totalAiIsmAdverbs
4
found
0
adverb
"lazily"
count
1
1
adverb
"perfectly"
count
1
2
adverb
"slightly"
count
1
3
adverb
"gently"
count
1
highlights
0
"lazily"
1
"perfectly"
2
"slightly"
3
"gently"
100.00%
AI-ism character names
Target: 0 AI-default names (17 tracked, −20% each)
codexExemptions
(empty)
found
(empty)
100.00%
AI-ism location names
Target: 0 AI-default location names (33 tracked, −20% each)
codexExemptions
(empty)
found
(empty)
54.92%
AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount
1442
totalAiIsms
13
found
0
word
"echoing"
count
1
1
word
"stark"
count
1
2
word
"gloom"
count
1
3
word
"etched"
count
1
4
word
"flicked"
count
1
5
word
"footsteps"
count
1
6
word
"echoed"
count
1
7
word
"traced"
count
2
8
word
"dancing"
count
1
9
word
"tracing"
count
1
10
word
"furrowed"
count
1
11
word
"scanning"
count
1
highlights
0
"echoing"
1
"stark"
2
"gloom"
3
"etched"
4
"flicked"
5
"footsteps"
6
"echoed"
7
"traced"
8
"dancing"
9
"tracing"
10
"furrowed"
11
"scanning"
100.00%
Cliché density
Target: ≤1 cliche(s) per 800-word window
totalCliches
1
maxInWindow
1
found
0
label
"eyes widened/narrowed"
count
1
highlights
0
"eyes narrowed"
100.00%
Emotion telling (show vs. tell)
Target: ≤3% sentences with emotion telling
emotionTells
0
narrationSentences
96
matches
(empty)
100.00%
Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount
0
hedgeCount
0
narrationSentences
96
filterMatches
(empty)
hedgeMatches
(empty)
100.00%
Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences
133
gibberishSentences
0
adjustedGibberishSentences
0
longSentenceCount
0
runOnParagraphCount
0
giantParagraphCount
0
wordSaladCount
0
repetitionLoopCount
0
controlTokenCount
0
repeatedSegmentCount
0
maxSentenceWordsSeen
43
ratio
0
matches
(empty)
100.00%
Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans
0
markdownWords
0
totalWords
1442
ratio
0
matches
(empty)
100.00%
Missing dialogue indicators (quotation marks)
Target: ≤10% speech attributions without quotation marks
totalAttributions
18
unquotedAttributions
0
matches
(empty)
42.54%
Name drop frequency
Target: ≤1.0 per-name mentions per 100 words
totalMentions
31
wordCount
791
uniqueNames
4
maxNameDensity
2.15
worstName
"Harlow"
maxWindowNameDensity
3
worstWindowName
"Harlow"
discoveredNames
Tube
1
Quinn
1
Harlow
17
Miller
12
persons
0
"Quinn"
1
"Harlow"
2
"Miller"
places
(empty)
globalScore
0.425
windowScore
0.667
100.00%
Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences
61
glossingSentenceCount
1
matches
0
"looked like the same style as the etching"
100.00%
"Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches
1
per1kWords
0.693
wordCount
1442
matches
0
"Not blood or decay, but a thick, greasy scent like a spent match"
100.00%
Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount
0
totalSentences
133
matches
(empty)
100.00%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
52
mean
27.73
std
20.13
cv
0.726
sampleLengths
0
65
1
11
2
76
3
12
4
21
5
40
6
16
7
5
8
84
9
18
10
19
11
49
12
26
13
41
14
3
15
38
16
14
17
49
18
36
19
3
20
10
21
43
22
37
23
2
24
4
25
9
26
9
27
50
28
30
29
47
30
15
31
21
32
34
33
23
34
38
35
45
36
32
37
11
38
5
39
23
40
10
41
13
42
10
43
31
44
58
45
40
46
22
47
7
48
70
49
3
86.99%
Passive voice overuse
Target: ≤2% passive sentences
passiveCount
5
totalSentences
96
matches
0
"was etched"
1
"were scrawled"
2
"was clutched"
3
"was drawn"
4
"was agitated"
58.16%
Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount
3
totalVerbs
141
matches
0
"was still dancing"
1
"wasn't spinning"
2
"was jerking"
100.00%
Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount
0
semicolonCount
0
flaggedSentences
0
totalSentences
133
ratio
0
matches
(empty)
99.69%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
793
adjectiveStacks
0
stackExamples
(empty)
adverbCount
32
adverbRatio
0.0403530895334174
lyAdverbCount
8
lyAdverbRatio
0.01008827238335435
100.00%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
133
echoCount
0
echoWords
(empty)
100.00%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
133
mean
10.84
std
8.35
cv
0.77
sampleLengths
0
4
1
25
2
14
3
22
4
11
5
11
6
15
7
15
8
11
9
3
10
5
11
16
12
6
13
3
14
3
15
10
16
11
17
11
18
13
19
7
20
9
21
9
22
7
23
4
24
1
25
16
26
8
27
22
28
38
29
8
30
5
31
5
32
6
33
13
34
12
35
14
36
11
37
12
38
12
39
14
40
3
41
8
42
30
43
3
44
4
45
11
46
23
47
2
48
12
49
13
52.13%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
9
diversityRatio
0.3533834586466165
totalSentences
133
uniqueOpeners
47
0.00%
Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount
0
totalSentences
87
matches
(empty)
ratio
0
49.89%
Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount
37
totalSentences
87
matches
0
"She gave a curt nod,"
1
"His face was a mask"
2
"Her knees popped."
3
"She ignored it."
4
"She flicked her wrist, but"
5
"She didn't look up."
6
"He smelled of cheap coffee"
7
"He gestured to the body"
8
"He nudged his chin at"
9
"She walked towards them, her"
10
"They looked like the same"
11
"Her eyes traced the line"
12
"she said again"
13
"She pointed her torch at"
14
"She knelt again, her eyes"
15
"She stood up and took"
16
"She stopped, ten feet from"
17
"He lumbered over."
18
"She pointed at the concrete."
19
"She traced a faint, smeared"
ratio
0.425
5.98%
Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount
79
totalSentences
87
matches
0
"The smell hit first."
1
"Water dripped somewhere in the"
2
"Harlow Quinn ducked under the"
3
"a constable called, his voice"
4
"She gave a curt nod,"
5
"The body lay crumpled on"
6
"A man, mid-thirties, dressed in"
7
"His face was a mask"
8
"A single item rested on"
9
"Harlow asked, crouching"
10
"Her knees popped."
11
"She ignored it."
12
"the constable said, jerking a"
13
"Harlow pulled on a latex"
14
"The face was etched with"
15
"The needle spun lazily, not"
16
"She flicked her wrist, but"
17
"A set of heavy footsteps"
18
"She didn't look up."
19
"DS Miller, a big man"
ratio
0.908
0.00%
Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount
0
totalSentences
87
matches
(empty)
ratio
0
100.00%
Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences
28
technicalSentenceCount
1
matches
0
"She walked towards them, her torchlight catching the crude, looping lines."
45.00%
Useless dialogue additions
Target: ≤5% dialogue tags with trailing filler fragments
totalTags
25
uselessAdditionCount
4
matches
0
"a constable called, his voice bouncing off the grimy tiles"
1
"Harlow said, more to herself than to Miller"
2
"Harlow said, her voice low, intense"
3
"Harlow said, her eyes scanning the dark"
100.00%
Dialogue tag variety (said vs. fancy)
Target: ≤10% fancy dialogue tags
totalTags
16
fancyCount
3
fancyTags
0
"Harlow murmured (murmur)"
1
"Harlow repeated (repeat)"
2
"Harlow admitted (admit)"
dialogueSentences
63
tagDensity
0.254
leniency
0.508
rawRatio
0.188
effectiveRatio
0.095
79.3808%