Run: 2026-02-17-01KHMPE4BNND0CZDWP85PZBRG7

Run Details

Model:
DeepSeek-V2 Chat
Test:
Bad Writing Habits
Scenario:
Mystery: examining a crime scene
Duration:
19.5s
Cost:
$0.001160
Provider:
Novita

Evaluation summary

Score Evaluator
Details Link
100.00% Adverbs in dialogue tags (details)
66.59% AI-ism adverb frequency (details)
100.00% AI-ism character names (details)
100.00% AI-ism location names (details)
55.46% 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)
28.79% Name drop frequency (details)
36.36% 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)
81.34% Passive voice overuse (details)
100.00% Past progressive (was/were + -ing) overuse (details)
0.00% Em-dash & semicolon overuse (details)
100.00% Purple prose (modifier overload) (details)
100.00% Repeated phrase echo (details)
100.00% Sentence length variance (details)
65.43% Sentence opener variety (details)
100.00% Adverb-first sentence starts (details)
100.00% Pronoun-first sentence starts (details)
75.38% Subject-first sentence starts (details)
100.00% Subordinate conjunction sentence starts (details)
23.81% Technical jargon density (details)
0.00% Useless dialogue additions (details)
100.00% Dialogue tag variety (said vs. fancy) (details)
81.1055% 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="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> 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.

Model response

AI
The abandoned Tube station smelled of damp concrete and something metallic underneath—blood, maybe, or rust from the disused tracks. Detective Harlow Quinn stepped carefully over the threshold, her boots crunching on broken glass. The emergency lights flickered overhead, casting long shadows across the platform where the body lay. Eva Kowalski stood near the edge, her round glasses catching the dim light as she adjusted them with one finger. Her freckled face was pale, her green eyes wide behind the lenses. She clutched her worn leather satchel to her chest like a shield. "You shouldn't be here," Quinn said, voice clipped. She didn't look at Eva, scanning the scene instead. The victim—male, mid-thirties, dark hair—was sprawled unnaturally, limbs twisted as if he'd been dropped from a height. But there was nothing above him but the vaulted ceiling, its tiles cracked and peeling. "I was already here," Eva said, tucking a loose curl behind her ear. "Research. The Market moves tonight." Quinn's jaw tightened. The Veil Market. She'd heard whispers of it—dealers selling things that shouldn't exist, smugglers trading in the impossible. She'd never seen it herself, but Morris had. Before he died. She crouched beside the body. No visible wounds. No blood pooling beneath him. Just that unnatural angle to his neck, like his spine had been snapped clean. She checked his pockets—empty. No wallet, no phone. But something glinted near his collar. She tugged it free. A small brass compass, its casing weathered with verdigris. The needle spun lazily, not settling on north. Eva sucked in a breath. "That's a Veil Compass." Quinn turned it over. The face was etched with strange symbols—protective sigils, maybe. "What does it do?" "It points to rifts. Places where the barriers between worlds are thin." Eva hesitated. "He was looking for something. Or running from it." Quinn stood, pocketing the compass. "You're telling me he was killed by something that doesn't exist." Eva's fingers tightened around her satchel strap. "I'm telling you the evidence doesn't fit a normal crime scene. No weapon. No signs of struggle. Just a broken neck and a missing Market token." Quinn frowned. "Token?" "The bone token. You need one to get into the Veil. He should have had it on him." Quinn scanned the platform again. No token. No witnesses. Just the flickering lights and the distant drip of water. And the body, posed like a warning. She exhaled sharply. "You think something came through. Something that didn't want him talking." Eva nodded. Quinn's fingers brushed the worn leather of her watch. Morris had been investigating something like this. Something that left no trace. She looked at Eva. "Show me the Market."
created_at"2026-02-17T02:21:48.414119+00:00" 2/17/2026, 2:21:48 AM
model"deepseek/deepseek-chat-v3"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency925
moderation_latency(null)
generation_time19448
tokens_prompt923
tokens_completion726
native_tokens_prompt856
native_tokens_completion629
native_tokens_completion_images(null)
native_tokens_reasoning0
native_tokens_cached0
num_media_prompt(null)
num_input_audio_prompt(null)
num_media_completion0
num_search_results(null)
origin"https://ncbench.com/"
is_byokfalse
finish_reason"stop"
native_finish_reason"stop"
usage0.0011601
router(null)
provider_responses
0
id"a2fc439787dcd8ee54d63c33adb4d08b"
status200
is_byokfalse
latency925
endpoint_id"cefb5a1e-9fea-4496-9a64-4e1b11a7cc8c"
provider_name"Novita"
model_permaslug"deepseek/deepseek-chat-v3"
api_type"completions"
id"gen-1771294888-RidLJ52KpkULjArJ8D1c"
upstream_id"a2fc439787dcd8ee54d63c33adb4d08b"
total_cost0.0011601
cache_discount(null)
upstream_inference_cost0
provider_name"Novita"

Evaluation details

Result Evaluator Details Meta Data
100.00% Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags3
adverbTagCount0
adverbTags(empty)
dialogueSentences13
tagDensity0.231
leniency0.462
rawRatio0
effectiveRatio0
66.59% AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount449
totalAiIsmAdverbs3
found
0
adverb"carefully"
count1
1
adverb"lazily"
count1
2
adverb"sharply"
count1
highlights
0"carefully"
1"lazily"
2"sharply"
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)
55.46% AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount449
totalAiIsms4
found
0
word"flickered"
count1
1
word"scanning"
count1
2
word"etched"
count1
3
word"scanned"
count1
highlights
0"flickered"
1"scanning"
2"etched"
3"scanned"
100.00% Cliché density
Target: ≤1 cliche(s) per 800-word window
totalCliches0
maxInWindow0
found(empty)
highlights(empty)
100.00% Emotion telling (show vs. tell)
Target: ≤3% sentences with emotion telling
emotionTells0
narrationSentences44
matches(empty)
100.00% Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount1
hedgeCount0
narrationSentences44
filterMatches
0"watch"
hedgeMatches(empty)
100.00% Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences54
gibberishSentences0
adjustedGibberishSentences0
longSentenceCount0
runOnParagraphCount0
giantParagraphCount0
wordSaladCount0
repetitionLoopCount0
controlTokenCount0
maxSentenceWordsSeen26
ratio0
matches(empty)
100.00% Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans0
markdownWords0
totalWords443
ratio0
matches(empty)
100.00% Missing dialogue indicators (quotation marks)
Target: ≤10% speech attributions without quotation marks
totalAttributions4
unquotedAttributions0
matches(empty)
28.79% Name drop frequency
Target: ≤1.0 per-name mentions per 100 words
totalMentions23
wordCount330
uniqueNames8
maxNameDensity2.42
worstName"Quinn"
maxWindowNameDensity3
worstWindowName"Quinn"
discoveredNames
Tube1
Harlow1
Quinn8
Kowalski1
Eva8
Veil1
Market1
Morris2
persons
0"Harlow"
1"Quinn"
2"Kowalski"
3"Eva"
4"Morris"
places(empty)
globalScore0.288
windowScore0.667
36.36% Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences22
glossingSentenceCount1
matches
0"something like this"
100.00% "Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches0
per1kWords0
wordCount443
matches(empty)
100.00% Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount0
totalSentences54
matches(empty)
100.00% Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs19
mean23.32
std14.39
cv0.617
sampleLengths
048
144
249
318
432
545
617
79
817
923
1016
1133
123
1318
1426
1514
162
1721
188
81.34% Passive voice overuse
Target: ≤2% passive sentences
passiveCount3
totalSentences44
matches
0"was sprawled"
1"been dropped"
2"been snapped"
3"was etched"
100.00% Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount0
totalVerbs59
matches(empty)
0.00% Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount6
semicolonCount0
flaggedSentences5
totalSentences54
ratio0.093
matches
0"The abandoned Tube station smelled of damp concrete and something metallic underneath—blood, maybe, or rust from the disused tracks."
1"The victim—male, mid-thirties, dark hair—was sprawled unnaturally, limbs twisted as if he'd been dropped from a height."
2"She'd heard whispers of it—dealers selling things that shouldn't exist, smugglers trading in the impossible."
3"She checked his pockets—empty."
4"The face was etched with strange symbols—protective sigils, maybe."
100.00% Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount337
adjectiveStacks0
stackExamples(empty)
adverbCount13
adverbRatio0.03857566765578635
lyAdverbCount4
lyAdverbRatio0.011869436201780416
100.00% Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences54
echoCount0
echoWords(empty)
100.00% Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences54
mean8.2
std5.48
cv0.668
sampleLengths
019
114
215
320
412
512
68
79
817
915
1013
115
123
133
1415
158
163
175
183
195
2014
214
224
236
244
259
268
275
284
294
309
314
3214
339
345
3511
367
3726
382
391
4018
415
422
432
4410
457
463
4711
482
499
65.43% Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats3
diversityRatio0.42592592592592593
totalSentences54
uniqueOpeners23
100.00% Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount2
totalSentences39
matches
0"Just that unnatural angle to"
1"Just the flickering lights and"
ratio0.051
100.00% Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount10
totalSentences39
matches
0"Her freckled face was pale,"
1"She clutched her worn leather"
2"She didn't look at Eva,"
3"She'd heard whispers of it—dealers"
4"She'd never seen it herself,"
5"She crouched beside the body."
6"She checked his pockets—empty."
7"She tugged it free."
8"She exhaled sharply."
9"She looked at Eva."
ratio0.256
75.38% Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount30
totalSentences39
matches
0"The abandoned Tube station smelled"
1"Detective Harlow Quinn stepped carefully"
2"The emergency lights flickered overhead,"
3"Eva Kowalski stood near the"
4"Her freckled face was pale,"
5"She clutched her worn leather"
6"Quinn said, voice clipped"
7"She didn't look at Eva,"
8"The victim—male, mid-thirties, dark hair—was"
9"Eva said, tucking a loose"
10"Quinn's jaw tightened."
11"The Veil Market."
12"She'd heard whispers of it—dealers"
13"She'd never seen it herself,"
14"She crouched beside the body."
15"She checked his pockets—empty."
16"She tugged it free."
17"A small brass compass, its"
18"The needle spun lazily, not"
19"Eva sucked in a breath."
ratio0.769
100.00% Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount1
totalSentences39
matches
0"Before he died."
ratio0.026
23.81% Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences12
technicalSentenceCount2
matches
0"The victim—male, mid-thirties, dark hair—was sprawled unnaturally, limbs twisted as if he'd been dropped from a height."
1"She'd heard whispers of it—dealers selling things that shouldn't exist, smugglers trading in the impossible."
0.00% Useless dialogue additions
Target: ≤5% dialogue tags with trailing filler fragments
totalTags3
uselessAdditionCount1
matches
0"Quinn said, voice clipped"
100.00% Dialogue tag variety (said vs. fancy)
Target: ≤10% fancy dialogue tags
totalTags2
fancyCount0
fancyTags(empty)
dialogueSentences13
tagDensity0.154
leniency0.308
rawRatio0
effectiveRatio0
81.1055%