Run: 2026-02-20-01KHY8WM2GF043NT62R49ZDRMZ

Run Details

Model:
Ministral 3 3B
Test:
Bad Writing Habits
Scenario:
Mystery: examining a crime scene
Duration:
2.6s
Cost:
$0.000173
Provider:
Mistral

Evaluation summary

Score Evaluator
Details Link
100.00% Adverbs in dialogue tags (details)
79.47% AI-ism adverb frequency (details)
100.00% AI-ism character names (details)
100.00% AI-ism location names (details)
0.00% AI-ism word frequency (details)
100.00% Cliché density (details)
100.00% Emotion telling (show vs. tell) (details)
57.14% Filter word density (details)
100.00% Gibberish response detection (details)
100.00% Markdown formatting overuse (details)
100.00% Missing dialogue indicators (quotation marks) (details)
11.11% 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)
100.00% Passive voice overuse (details)
100.00% Past progressive (was/were + -ing) overuse (details)
0.00% Em-dash & semicolon overuse (details)
98.79% Purple prose (modifier overload) (details)
100.00% Repeated phrase echo (details)
100.00% Sentence length variance (details)
45.03% Sentence opener variety (details)
0.00% Adverb-first sentence starts (details)
100.00% Pronoun-first sentence starts (details)
13.19% Subject-first sentence starts (details)
0.00% Subordinate conjunction sentence starts (details)
100.00% Technical jargon density (details)
0.00% Useless dialogue additions (details)
73.08% Dialogue tag variety (said vs. fancy) (details)
72.5937% 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 flickering neon glow of the abandoned Tube station pulsed like a dying heartbeat. Detective Harlow Quinn stepped through the rusted metal doors, her boots crunching on broken glass. The air smelled of damp earth and something older—something that didn’t belong here. A body lay sprawled across the platform, its face twisted in a silent scream. The victim’s throat had been slit, but the blood had pooled too neatly, too *wrong*. The cuts weren’t sharp enough for a blade. The edges were jagged, like something had torn through flesh. “Quinn,” a voice cut through the silence. She turned to see DS Morris—no, not DS Morris. Her partner, gone three years ago in a case that had left her with more questions than answers. But here she stood, her uniform still crisp, her expression unreadable. “You’re late.” Quinn didn’t answer. She knelt beside the body, her fingers brushing the damp fabric. The wound wasn’t deep enough to kill. Not like this. Not cleanly. The victim’s eyes were open, staring blankly at the ceiling. Too still. “You’re staring at the eyes,” Morris said, her voice low. “Like you’re waiting for something.” Quinn didn’t look up. “What do you see?” Morris exhaled sharply, her breath warm against the cold. “A man who wasn’t supposed to be here. Not in this place.” A distant hum vibrated through the station, like a machine turning on beneath the floorboards. Quinn’s gaze flicked toward the shadows where the walls seemed to breathe. The Veil Market moved every full moon, but this wasn’t the usual chaos of deals and whispers. This was too quiet. “You’re not supposed to be here,” Quinn muttered. Morris didn’t deny it. “I know.” The station groaned as something heavy shifted beneath them. Quinn’s pulse quickened. The needle on her watch spun, the second hand moving too fast. She reached for her radio, but her fingers brushed against something cold and smooth—brass, etched with sigils. “What’s that?” Morris asked, her voice tight. Quinn’s fingers closed around the object. The compass in her palm hummed, its needle pointing straight at the victim’s chest. Not at the body. At the spot where the wound should have been. “This isn’t right,” Quinn said, her voice barely above a whisper. Morris didn’t argue. She just stood there, her eyes locked on the compass, her jaw set. “We need to get out of here.” The station’s lights flickered. The hum grew louder, a low, rhythmic pulse that made Quinn’s skin prickle. The walls seemed to close in, the air thick with something unseen. “Morris,” Quinn said, her voice urgent. “What’s happening?” Morris didn’t answer. She just turned, her back to Quinn, and walked toward the exit. The doors hissed open, revealing the dim glow of the underground tunnels. Quinn followed, her breath ragged, her mind racing. The Veil Market wasn’t just a market. It was a prison. And they’d just walked into it.
created_at"2026-02-20T19:37:07.421Z" 2/20/2026, 7:37:07 PM
model"mistralai/ministral-3b-2512"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency213
moderation_latency(null)
generation_time2186
tokens_prompt1149
tokens_completion791
native_tokens_prompt1062
native_tokens_completion664
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"
usage0.0001726
router(null)
provider_responses
0
endpoint_id"b9f6828f-aef9-498a-91f5-6816fbf72420"
id"0bf38f9d99c94a4cb6f8d919d44601cc"
is_byokfalse
latency213
model_permaslug"mistralai/ministral-3b-2512"
provider_name"Mistral"
status200
api_type"completions"
id"gen-1771616227-4B84dUBWiC1lWBou1mct"
upstream_id"0bf38f9d99c94a4cb6f8d919d44601cc"
total_cost0.0001726
cache_discount(null)
upstream_inference_cost0
provider_name"Mistral"

Evaluation details

Result Evaluator Details Meta Data
100.00% Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags5
adverbTagCount0
adverbTags(empty)
dialogueSentences13
tagDensity0.385
leniency0.769
rawRatio0
effectiveRatio0
79.47% AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount487
totalAiIsmAdverbs2
found
0
adverb"sharply"
count1
1
adverb"barely above a whisper"
count1
highlights
0"sharply"
1"barely above a whisper"
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)
0.00% AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount487
totalAiIsms13
found
0
word"pulsed"
count1
1
word"silence"
count1
2
word"unreadable"
count1
3
word"vibrated"
count1
4
word"flicked"
count1
5
word"pulse"
count2
6
word"quickened"
count1
7
word"etched"
count1
8
word"whisper"
count1
9
word"flickered"
count1
10
word"rhythmic"
count1
11
word"racing"
count1
highlights
0"pulsed"
1"silence"
2"unreadable"
3"vibrated"
4"flicked"
5"pulse"
6"quickened"
7"etched"
8"whisper"
9"flickered"
10"rhythmic"
11"racing"
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
narrationSentences50
matches(empty)
57.14% Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount1
hedgeCount2
narrationSentences50
filterMatches
0"watch"
hedgeMatches
0"seemed to"
100.00% Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences57
gibberishSentences0
adjustedGibberishSentences0
longSentenceCount0
runOnParagraphCount0
giantParagraphCount0
wordSaladCount0
repetitionLoopCount0
controlTokenCount0
maxSentenceWordsSeen18
ratio0
matches(empty)
100.00% Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans1
markdownWords1
totalWords484
ratio0.002
matches
0"wrong"
100.00% Missing dialogue indicators (quotation marks)
Target: ≤10% speech attributions without quotation marks
totalAttributions9
unquotedAttributions0
matches(empty)
11.11% Name drop frequency
Target: ≤1.0 per-name mentions per 100 words
totalMentions26
wordCount432
uniqueNames6
maxNameDensity2.78
worstName"Quinn"
maxWindowNameDensity4
worstWindowName"Quinn"
discoveredNames
Tube1
Harlow1
Quinn12
Morris8
Veil2
Market2
persons
0"Harlow"
1"Quinn"
2"Morris"
3"Market"
places(empty)
globalScore0.111
windowScore0.333
100.00% Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences33
glossingSentenceCount0
matches(empty)
100.00% "Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches0
per1kWords0
wordCount484
matches(empty)
100.00% Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount0
totalSentences57
matches(empty)
100.00% Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs19
mean25.47
std15.08
cv0.592
sampleLengths
042
147
247
338
415
58
621
748
88
96
1041
117
1233
1311
1423
1529
168
1735
1817
100.00% Passive voice overuse
Target: ≤2% passive sentences
passiveCount0
totalSentences50
matches(empty)
100.00% Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount0
totalVerbs70
matches(empty)
0.00% Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount3
semicolonCount0
flaggedSentences3
totalSentences57
ratio0.053
matches
0"The air smelled of damp earth and something older—something that didn’t belong here."
1"She turned to see DS Morris—no, not DS Morris."
2"She reached for her radio, but her fingers brushed against something cold and smooth—brass, etched with sigils."
98.79% Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount435
adjectiveStacks0
stackExamples(empty)
adverbCount18
adverbRatio0.041379310344827586
lyAdverbCount5
lyAdverbRatio0.011494252873563218
100.00% Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences57
echoCount0
echoWords(empty)
100.00% Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences57
mean8.49
std4.51
cv0.531
sampleLengths
014
115
213
314
415
58
610
77
89
918
1011
112
123
1311
147
153
162
1710
182
1910
205
214
224
239
2412
2515
2612
2717
284
298
304
312
329
333
3412
3517
367
376
3814
394
409
4111
423
4313
447
454
4613
4712
486
492
45.03% Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats6
diversityRatio0.3333333333333333
totalSentences57
uniqueOpeners19
0.00% Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount0
totalSentences47
matches(empty)
ratio0
100.00% Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount7
totalSentences47
matches
0"She turned to see DS"
1"Her partner, gone three years"
2"She knelt beside the body,"
3"She reached for her radio,"
4"She just stood there, her"
5"She just turned, her back"
6"It was a prison."
ratio0.149
13.19% Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount42
totalSentences47
matches
0"The flickering neon glow of"
1"Detective Harlow Quinn stepped through"
2"The air smelled of damp"
3"A body lay sprawled across"
4"The victim’s throat had been"
5"The cuts weren’t sharp enough"
6"The edges were jagged, like"
7"a voice cut through the"
8"She turned to see DS"
9"Her partner, gone three years"
10"Quinn didn’t answer."
11"She knelt beside the body,"
12"The wound wasn’t deep enough"
13"The victim’s eyes were open,"
14"Morris said, her voice low"
15"Quinn didn’t look up."
16"Morris exhaled sharply, her breath"
17"A distant hum vibrated through"
18"Quinn’s gaze flicked toward the"
19"The Veil Market moved every"
ratio0.894
0.00% Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount0
totalSentences47
matches(empty)
ratio0
100.00% Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences22
technicalSentenceCount1
matches
0"The air smelled of damp earth and something older—something that didn’t belong here."
0.00% Useless dialogue additions
Target: ≤5% dialogue tags with trailing filler fragments
totalTags5
uselessAdditionCount4
matches
0"Morris said, her voice low"
1"Morris asked, her voice tight"
2"Quinn said, her voice barely above a whisper"
3"Quinn said, her voice urgent"
73.08% Dialogue tag variety (said vs. fancy)
Target: ≤10% fancy dialogue tags
totalTags5
fancyCount1
fancyTags
0"Quinn muttered (mutter)"
dialogueSentences13
tagDensity0.385
leniency0.769
rawRatio0.2
effectiveRatio0.154
72.5937%