Run: 2026-02-19-01KHTY139NZ6C900FW03925JEC

Run Details

Model:
DeepSeek V3 (2025-03-24)
Test:
Bad Writing Habits
Scenario:
Mystery: examining a crime scene
Duration:
14.5s
Cost:
$0.000832
Provider:
ModelRun

Evaluation summary

Score Evaluator
Details Link
100.00% Adverbs in dialogue tags (details)
89.71% AI-ism adverb frequency (details)
100.00% AI-ism character names (details)
100.00% AI-ism location names (details)
38.27% AI-ism word frequency (details)
100.00% Cliché density (details)
100.00% Emotion telling (show vs. tell) (details)
100.00% Filter word density (details)
0.00% Gibberish response detection (details)
100.00% Markdown formatting overuse (details)
100.00% Missing dialogue indicators (quotation marks) (details)
0.00% 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)
80.20% 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)
98.21% Sentence opener variety (details)
95.24% Adverb-first sentence starts (details)
100.00% Pronoun-first sentence starts (details)
31.43% Subject-first sentence starts (details)
0.00% Subordinate conjunction sentence starts (details)
100.00% Technical jargon density (details)
75.00% Useless dialogue additions (details)
100.00% Dialogue tag variety (said vs. fancy) (details)
80.2688% 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

AI
The stale air clung to Detective Harlow Quinn’s throat as she descended the metal staircase into the abandoned Tube station, her boots echoing against rusted steps. Black-and-yellow tape cordoned่วน off the platform of The Veil Market, uniformed officers milling like nervous shadows. Eva Kowalski stood near the edge, her freckled face illuminated by the wavering glow of an emergency lantern. She clutched the straps of her satchel, fingers white-knuckled. "You took your time." Quinn checked her watch—too slow for Eva’s sarcasm. "Traffic." She stepped closer, scanning the scene. Broken vials littered the tiles, their contents soaked into the cracks like spilled ink. The stall behind Eva was torn apart, its curtains shredded, wooden beams splintered inward. "CSU’s already been through," Eva muttered, tucking a curl behind her ear. "No prints. No fibers." Quinn crouched near the largest shard of glass, tilting her head. "Blood?" "Some. Not much." The jagged edges reflected the dim light in odd patterns, like liquid mercury. Too clean for a struggle. "This wasn’t a robbery." Eva’s laugh was sharp. "Oh, enlighten me, Detective." Quinn ignored her, moving toward the back of the stall. A brass compass lay discarded beneath a toppled shelf, its patina catching the light. She nudged it with her shoe. "Yours?" Eva stiffened. "What?" "The Veil Compass." Quinn didn’t look up. "Assmarks." "I—that’s a collector’s item." Eva’s voice pitched higher. "They sell them here." "Maybe." Quinn knelt, careful not to touch it. The sigils on its face were worn with use. The needle quivered, pointing northeast—toward the tunnel where the tracks vanished into shadow. She exhaled through her nose. "Where’s the victim?" Eva hesitated. "There wasn’t one." Quinn rose. "People don’t tear places apart "Chewed out of boredom." "The Market moves locations every full moon." Eva hugged her satchel tighter. "Maybe sidious did you Due根本理由# load_job"></error>硬. "New tables, drapes. Business the blotter." Quinn studied the西门布林幕, tracing the where rivets had been wrenched from the frame. The splinters clung outward—from the inside. She turned slowly. "You’re telling me someone broke in. Tore elbows through this stall. Left empty bottles. Sanky也成为曾肃." Her jaw set. "And took nothing?" Eva’s greensú fired违敵视. "Think you’re sumbox think美 ten steps ahead—" "Then help me understand." Quinn stepped in close, voice lowering. "Because tonight台い, someone went to a hell of a lot of trouble to fake a crime scene." Eva’s breath hitched. The compass needle p티레 3/4 of the way toward the tunnel before settling back with a click. Quinn’s gaze didn’t waver. "Tell me what I’m missing." A drop of sweat traced Eva’s temple. She opened her mouth— Boots thudded on the platform. A uniformed officer approached, radio squawking. "Ma’am, backup’s arrived." Eva looked away first. Quinn retrieved the compass毕uss%" and pocketed it without a word. "Hey," Evai圢 began. "That’s evidence—" "Yeah." Quinn surveyed the tunnel’s mouth, where the shadows pooled too thick, too dark. "Sure it is."
created_at"2026-02-19T12:29:36.448Z" 2/19/2026, 12:29:36 PM
model"deepseek/deepseek-chat-v3-0324"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency591
moderation_latency(null)
generation_time14501
tokens_prompt1149
tokens_completion873
native_tokens_prompt1032
native_tokens_completion812
native_tokens_completion_images(null)
native_tokens_reasoning0
native_tokens_cached4
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"
usage0.00083164
router(null)
provider_responses
0
endpoint_id"a3938710-1b99-4a1f-85f4-2494f3f659d3"
id"chatcmpl-09774050da804f6db61f1f547a6f36da"
is_byokfalse
latency278
model_permaslug"deepseek/deepseek-chat-v3-0324"
provider_name"ModelRun"
status200
api_type"completions"
id"gen-1771504176-TAVu73emt3ADGpXts38P"
upstream_id"chatcmpl-09774050da804f6db61f1f547a6f36da"
total_cost0.00083164
cache_discount(null)
upstream_inference_cost0
provider_name"ModelRun"

Evaluation details

Result Evaluator Details Meta Data
100.00% Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags13
adverbTagCount0
adverbTags(empty)
dialogueSentences31
tagDensity0.419
leniency0.839
rawRatio0
effectiveRatio0
89.71% AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount486
totalAiIsmAdverbs1
found
0
adverb"slowly"
count1
highlights
0"slowly"
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)
38.27% AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount486
totalAiIsms6
found
0
word"echoing"
count1
1
word"wavering"
count1
2
word"scanning"
count1
3
word"quivered"
count1
4
word"tracing"
count1
5
word"traced"
count1
highlights
0"echoing"
1"wavering"
2"scanning"
3"quivered"
4"tracing"
5"traced"
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
narrationSentences42
matches(empty)
100.00% Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount1
hedgeCount0
narrationSentences42
filterMatches
0"watch"
hedgeMatches(empty)
0.00% Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences56
gibberishSentences9
adjustedGibberishSentences9
longSentenceCount0
runOnParagraphCount0
giantParagraphCount0
wordSaladCount0
repetitionLoopCount0
controlTokenCount0
maxSentenceWordsSeen29
ratio0.161
matches
0"Black-and-yellow tape cordoned่วน off the platform of The Veil Market, uniformed officers milling like nervous shadows."
1"\"The Market moves locations every full moon.\" Eva hugged her satchel tighter. \"Maybe sidious did you Due根本理由# load_job\"></error>硬. \"New tables, drapes."
2"Quinn studied the西门布林幕, tracing the where rivets had been wrenched from the frame. The splinters clung outward—from the inside. She turned slowly. \"You’re telling me someone broke …"
3"Sanky也成为曾肃.\" Her jaw set. \"And took nothing?\""
4"Eva’s greensú fired违敵视. \"Think you’re sumbox think美 ten steps ahead—\""
5"\"Then help me understand.\" Quinn stepped in close, voice lowering. \"Because tonight台い, someone went to a hell of a lot of trouble to fake a crime scene.\""
6"Eva’s breath hitched. The compass needle p티레 3/4 of the way toward the tunnel before settling back with a click."
7"Quinn retrieved the compass毕uss%\" and pocketed it without a word."
8"\"Hey,\" Evai圢 began."
100.00% Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans0
markdownWords0
totalWords479
ratio0
matches(empty)
100.00% Missing dialogue indicators (quotation marks)
Target: ≤10% speech attributions without quotation marks
totalAttributions3
unquotedAttributions0
matches(empty)
0.00% Name drop frequency
Target: ≤1.0 per-name mentions per 100 words
totalMentions34
wordCount349
uniqueNames11
maxNameDensity3.72
worstName"Eva"
maxWindowNameDensity4.5
worstWindowName"Quinn"
discoveredNames
Detective1
Harlow1
Quinn12
Tube1
Veil1
Market1
Kowalski1
Eva13
Broken1
Business1
Boots1
persons
0"Harlow"
1"Quinn"
2"Market"
3"Kowalski"
4"Eva"
5"Boots"
places
0"Detective"
globalScore0
windowScore0.167
100.00% Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences24
glossingSentenceCount0
matches(empty)
100.00% "Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches0
per1kWords0
wordCount479
matches(empty)
100.00% Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount0
totalSentences56
matches(empty)
100.00% Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs28
mean17.11
std12.11
cv0.708
sampleLengths
042
131
243
316
412
53
622
78
831
93
108
1112
1230
138
145
1511
1624
1743
1810
1927
2020
219
2211
2314
244
2510
265
2717
80.20% Passive voice overuse
Target: ≤2% passive sentences
passiveCount3
totalSentences42
matches
0"was torn"
1"were worn"
2"been wrenched"
100.00% Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount0
totalVerbs64
matches(empty)
0.00% Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount4
semicolonCount0
flaggedSentences4
totalSentences56
ratio0.071
matches
0"Quinn checked her watch—too slow for Eva’s sarcasm."
1"The needle quivered, pointing northeast—toward the tunnel where the tracks vanished into shadow."
2"Quinn studied the西门布林幕, tracing the where rivets had been wrenched from the frame. The splinters clung outward—from the inside. She turned slowly. \"You’re telling me someone broke in."
3"A drop of sweat traced Eva’s temple. She opened her mouth—"
100.00% Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount336
adjectiveStacks0
stackExamples(empty)
adverbCount7
adverbRatio0.020833333333333332
lyAdverbCount1
lyAdverbRatio0.002976190476190476
100.00% Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences56
echoCount0
echoWords(empty)
100.00% Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences56
mean8.55
std6.66
cv0.779
sampleLengths
026
116
218
39
44
58
67
714
814
912
104
1111
121
133
1413
155
164
174
184
1910
2014
216
221
232
241
257
261
278
284
298
309
3113
325
333
342
353
362
379
3821
393
4028
415
423
437
4410
4527
4620
479
4811
4914
98.21% Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats1
diversityRatio0.625
totalSentences56
uniqueOpeners35
95.24% Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount1
totalSentences35
matches
0"Too clean for a struggle."
ratio0.029
100.00% Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount4
totalSentences35
matches
0"She clutched the straps of"
1"She stepped closer, scanning the"
2"She nudged it with her"
3"She exhaled through her nose."
ratio0.114
31.43% Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount30
totalSentences35
matches
0"The stale air clung to"
1"Eva Kowalski stood near the"
2"She clutched the straps of"
3"Quinn checked her watch—too slow"
4"She stepped closer, scanning the"
5"The stall behind Eva was"
6"Eva muttered, tucking a curl"
7"Quinn crouched near the largest"
8"The jagged edges reflected the"
9"Eva’s laugh was sharp."
10"Quinn ignored her, moving toward"
11"A brass compass lay discarded"
12"She nudged it with her"
13"Quinn didn’t look up"
14"Eva’s voice pitched higher"
15"Quinn knelt, careful not to"
16"The sigils on its face"
17"The needle quivered, pointing northeast—toward"
18"She exhaled through her nose."
19"Eva hugged her satchel tighter."
ratio0.857
0.00% Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount0
totalSentences35
matches(empty)
ratio0
100.00% Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences13
technicalSentenceCount0
matches(empty)
75.00% Useless dialogue additions
Target: ≤5% dialogue tags with trailing filler fragments
totalTags10
uselessAdditionCount1
matches
0"Quinn stepped, voice lowering"
100.00% Dialogue tag variety (said vs. fancy)
Target: ≤10% fancy dialogue tags
totalTags1
fancyCount1
fancyTags
0"Eva muttered (mutter)"
dialogueSentences31
tagDensity0.032
leniency0.065
rawRatio1
effectiveRatio0.065
80.2688%