NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-17-01KHNFW1984ZGYYA3RKHTTPMVW
Run Details
Model:
Arcee AI: Trinity Mini
Test:
Bad Writing Habits
Scenario:
Romance: separated couple reunites
Duration:
9.3s
Cost:
$0.000335
Provider:
Clarifai
Evaluation summary
Score
Evaluator
Details Link
100.00%
Adverbs in dialogue tags
(details)
71.80%
AI-ism adverb frequency
(details)
100.00%
AI-ism character names
(details)
100.00%
AI-ism location names
(details)
10.71%
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)
100.00%
Name drop frequency
(details)
95.65%
Narrator intent-glossing
(details)
11.14%
"Not X but Y" pattern overuse
(details)
100.00%
Overuse of "that" (subordinate clause padding)
(details)
87.60%
Paragraph length variance
(details)
92.02%
Passive voice overuse
(details)
100.00%
Past progressive (was/were + -ing) overuse
(details)
49.18%
Em-dash & semicolon overuse
(details)
87.93%
Purple prose (modifier overload)
(details)
100.00%
Repeated phrase echo
(details)
100.00%
Sentence length variance
(details)
60.66%
Sentence opener variety
(details)
100.00%
Adverb-first sentence starts
(details)
23.92%
Pronoun-first sentence starts
(details)
18.82%
Subject-first sentence starts
(details)
98.04%
Subordinate conjunction sentence starts
(details)
100.00%
Technical jargon density
(details)
0.00%
Useless dialogue additions
(details)
38.89%
Dialogue tag variety (said vs. fancy)
(details)
78.2127%
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="4412" name="Aurora Carter" aliases="Rory, Laila, Aurora, Carter, Malphora" pronouns="she/her" species="Human" storyRole="Protagonist"> <eyeColor>bright blue</eyeColor> <hair>straight shoulder-length black</hair> <height>5'6"</height> <age>25</age> <scar>small crescent-shaped scar on her left wrist from a childhood accident</scar> <description>Aurora, referred to by her friends as Rory, is cool-headed, intelligent, and known for quick out-of-the-box thinking.</description> <backstory>Born in Cardiff to an Irish barrister father (Brendan Carter) and Welsh teacher mother (Jennifer Carter née Ellis), Rory reluctantly pursued Pre-Law at Cardiff University. She fled to London at the behest of her childhood friend Eva after leaving an abusive ex named Evan. By day, Rory works part-time as a delivery person for Yu-Fei Cheung's Golden Empress restaurant. She lives in a flat above Silas' bar.</backstory> </character> <character id="4415" name="Lucien Moreau" aliases="Lucien, Luc, the Frenchman" pronouns="he/him" species="Half-Demon" storyRole="Ally"> <eyeColor>heterochromatic — one amber, one black</eyeColor> <hair>slicked-back platinum blond</hair> <height>5'11"</height> <age>32</age> <cane>ivory-handled cane that conceals a thin blade</cane> <dress>tailored charcoal suits</dress> <description>An impeccably dressed fixer and information broker in London's supernatural underworld. He speaks four languages fluently.</description> <backstory>Born in Marseille to a human mother and a demon father from the realm of Avaros.</backstory> </character> <location id="858" name="Eva's Flat" aliases="Eva's place, the flat on Brick Lane" realm="Earth"> <cat>tabby cat named Ptolemy</cat> <security>three deadbolts on the front door</security> <description>A cramped one-bedroom flat above a curry house on Brick Lane, East London. Every surface is covered with books, scrolls, and research notes.</description> </location> </codex> Write a fully fleshed out scene of at least 1000 words in the romance genre. Two people who parted on complicated terms are forced together again when one shows up unannounced at the other's flat. There's history between them — attraction, hurt, and things left unsaid. Write the scene from the moment the door opens. 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
The smell of curry andstale
beer hit Aurora the moment she pushed open the heavy door to her flat above Silas’s bar. Her flat was a cramped,
chaotic
sanctuary
, a single bedroom stacked with books, scrolls, and research notes that spilled onto every surface like a landslide of knowledge. Three deadbolts, each reinforced with a heavy chain, secured the front door, a silent
testament
to the fear that still clung to her like a second skin.
She’d fled Cardiff, not just
Evan, but the ghosts of her past, and this flat, above the raucous bar and the pungent curry house, felt like the safest place she’d ever known.
The tabby cat, Ptolemy, a
sleek, indifferent creature,
was sprawled
across her worn leather armchair, blinking
slowly
at the intrusion. “Ptolemy, you’re not helping,”
she murmured, stepping inside, the
door thudding shut behind her.
She’d just finished a late
shift delivering for Yu-Fei Cheung’s Golden Empress, the smell of spices still clinging to her clothes. She’d planned a quiet evening: a bath, a book, and maybe some much-needed sleep.
Instead, the flat felt suddenly
too small, the air thick with the
unspoken
weight
of the past.
She’d thought she’d left all
that behind when she’d fled Evan, the abusive ex who’d left that crescent-shaped scar on her left wrist. But here he was,
not Evan, but someone else, someone who’d been part of a different kind of
chaos. Lucien Moreau.
The name echoed in her
mind, a sharp, discordant note.
They’d met in the tangled
web of London’s supernatural underworld, a world she’d stumbled into while trying to escape her own.
He’d been a fixer, impeccably
dressed in charcoal suits, speaking four languages with unnerving fluency.
He’d helped her, once, but
there’d been a cost, a betrayal that still stung.
She’d thought she’d seen the
last of him when she’d left London for Cardiff, only to return months later, drawn back by Eva’s frantic call about something... something dangerous.
Now, he stood on her
doorstep, the ivory-handled cane tapping a precise rhythm on the worn wooden step.
His platinum blond hair was
slicked back, sharp angles framing his face, one amber eye and one black eye fixed on her with unnerving
intensity
. The scent of expensive cologne, sharp and clean, cut through the curry and beer.
He didn’t look like he
belonged in this cramped, book-filled flat, but then, he never had. “Aurora,” he said, his voice smooth, low, carrying an accent that was undeniably French. “Or should I call you Rory? Laila? Malphora?”
The names rolled off his
tongue like a practiced incantation, each one a different facet of her
fractured
identity. “I know you’re here. Eva told me. She’s worried. About you. About everything.” She didn’t move, didn’t speak, just watched him as he pushed the door open wider, letting the cool night air and the distant sound of Silas’s bar chatter seep in.
Ptolemy, sensing a stranger, hissed
and scrambled off the chair, vanishing under the sofa. The
silence
stretched, thick and
charged
, filled only by the distant clang of a pot in the curry house below. “You shouldn’t be here,”
she finally said, her voice
cool, controlled, though her heart hammered against her ribs like a trapped bird. “This isn’t a place for... for your kind.”
He chuckled, a low, dry
sound. “My kind? Aurora, you know as well as I do that this city is a patchwork quilt of many kinds. Humans, demons, things in between. And you, my dear, are
very
much a part of it now.” He took a step closer, his polished shoes clicking on the worn floorboards. “You’ve changed. Cardiff suits you, the law degree abandoned, the delivery job. But the fire’s still there. The quick thinking. The... resilience.”
He paused, his amber eye
studying her, then
flicked
to the scar on her wrist, a faint, pale crescent moon against her skin. “Evan’s gone, I assume? The abuse ended?” The question
hung in the air
, sharp and unexpected, cutting through the
carefully
constructed
facade
of calm. “He’s gone,”
she replied, her voice tight
. “He’s irrelevant now.”
She forced herself to meet
his gaze, to hold it steady. “What do you want, Lucien? Why are you here?” He didn’t answer directly. Instead, he lifted his cane, tapping it
lightly
against the floor. “I need your help, Rory. Eva’s in trouble. Deep trouble. Something ancient, something... hungry, is stirring in the shadows of this city. Something that even my connections can’t fully grasp. And she’s caught in its path.” The words landed like stones. Eva, her childhood friend, dragged back into the nightmare she’d fled. The ancient hunger, the supernatural threat – it all tied back to the world Lucien inhabited, the world she’d
tried to
escape. “Why me?”
she demanded, the cool mask slipping slightly
. “Why not your usual network of informants?” He smiled, a ghost of a smile that didn’t reach his eyes. “Because, Aurora, you’re the only one who can see it. The only one who can
navigate
the human world and the supernatural with equal ease. You’re the key, the wildcard. And Eva... she’s always been impulsive. She’s dug herself into a hole, and now she needs someone who knows how to climb out.” He took another step, closing the distance between them. The scent of his cologne mingled with the curry, a potent, unfamiliar mix. “I know we parted badly. I know I hurt you. But this... this is bigger than us. Bigger than the past.” His black eye held hers,
unwavering
. “Can you look past it? For Eva? For London?” The question hung between them, heavy with
unspoken
history and the
weight
of the impending danger. The scar on her wrist
pulsed
faintly, a
reminder
of the pain she’d endured, but also of her survival. The fire Lucien mentioned still burned within her, a fierce, protective instinct. Eva was in trouble. London was in danger. And Lucien Moreau, the half-demon fixer,
was offering
a lifeline, however tangled and painful the path might be. The choice wasn’t just about him; it was about everything she’d fought to build, everything she’d lost, and everything she might lose again. The door to her past had opened, and standing on the threshold was the man who’d been part of both the pain and the salvation. The decision, as always, rested with her.
created_at
"2026-02-17T09:46:08.054626+00:00"
2/17/2026, 9:46:08 AM
model
"arcee-ai/trinity-mini-20251201"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
76
moderation_latency
(null)
generation_time
9197
tokens_prompt
808
tokens_completion
2384
native_tokens_prompt
721
native_tokens_completion
2014
native_tokens_completion_images
(null)
native_tokens_reasoning
746
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
0.000334545
router
(null)
provider_responses
0
id
"974e8d3d49fb4885b5f2b3701a5c0cdc"
status
200
is_byok
false
latency
76
endpoint_id
"73d82c49-90ae-4223-a860-83ed0810821f"
provider_name
"Clarifai"
model_permaslug
"arcee-ai/trinity-mini-20251201"
api_type
"completions"
id
"gen-1771321558-12WyAsCiJMTskknA6Mx8"
upstream_id
"974e8d3d49fb4885b5f2b3701a5c0cdc"
total_cost
0.000334545
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Clarifai"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
10
adverbTagCount
1
adverbTags
0
"she finally said [finally]"
dialogueSentences
18
tagDensity
0.556
leniency
1
rawRatio
0.1
effectiveRatio
0.1
71.80%
AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount
1064
totalAiIsmAdverbs
6
found
0
adverb
"slowly"
count
1
1
adverb
"suddenly"
count
1
2
adverb
"very"
count
1
3
adverb
"carefully"
count
1
4
adverb
"lightly"
count
1
5
adverb
"slightly"
count
1
highlights
0
"slowly"
1
"suddenly"
2
"very"
3
"carefully"
4
"lightly"
5
"slightly"
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)
10.71%
AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount
1064
totalAiIsms
19
found
0
word
"chaotic"
count
1
1
word
"sanctuary"
count
1
2
word
"testament"
count
1
3
word
"unspoken"
count
2
4
word
"weight"
count
2
5
word
"echoed"
count
1
6
word
"intensity"
count
1
7
word
"fractured"
count
1
8
word
"silence"
count
1
9
word
"charged"
count
1
10
word
"flicked"
count
1
11
word
"constructed"
count
1
12
word
"facade"
count
1
13
word
"navigate"
count
1
14
word
"unwavering"
count
1
15
word
"pulsed"
count
1
16
word
"reminder"
count
1
highlights
0
"chaotic"
1
"sanctuary"
2
"testament"
3
"unspoken"
4
"weight"
5
"echoed"
6
"intensity"
7
"fractured"
8
"silence"
9
"charged"
10
"flicked"
11
"constructed"
12
"facade"
13
"navigate"
14
"unwavering"
15
"pulsed"
16
"reminder"
100.00%
Cliché density
Target: ≤1 cliche(s) per 800-word window
totalCliches
1
maxInWindow
1
found
0
label
"hung in the air"
count
1
highlights
0
"hung in the air"
100.00%
Emotion telling (show vs. tell)
Target: ≤3% sentences with emotion telling
emotionTells
0
narrationSentences
53
matches
(empty)
100.00%
Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount
0
hedgeCount
1
narrationSentences
53
filterMatches
(empty)
hedgeMatches
0
"tried to"
100.00%
Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences
61
gibberishSentences
0
adjustedGibberishSentences
0
longSentenceCount
0
runOnParagraphCount
0
giantParagraphCount
0
wordSaladCount
0
repetitionLoopCount
0
controlTokenCount
0
maxSentenceWordsSeen
53
ratio
0
matches
(empty)
100.00%
Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans
0
markdownWords
0
totalWords
1059
ratio
0
matches
(empty)
100.00%
Missing dialogue indicators (quotation marks)
Target: ≤10% speech attributions without quotation marks
totalAttributions
10
unquotedAttributions
0
matches
(empty)
100.00%
Name drop frequency
Target: ≤1.0 per-name mentions per 100 words
totalMentions
27
wordCount
813
uniqueNames
14
maxNameDensity
0.49
worstName
"Lucien"
maxWindowNameDensity
1.5
worstWindowName
"Evan"
discoveredNames
Aurora
1
Silas
2
Cardiff
2
Evan
3
Ptolemy
2
Yu-Fei
1
Cheung
1
Golden
1
Empress
1
Moreau
2
London
3
Eva
3
French
1
Lucien
4
persons
0
"Silas"
1
"Evan"
2
"Ptolemy"
3
"Yu-Fei"
4
"Cheung"
5
"Moreau"
6
"Eva"
7
"Lucien"
places
0
"Cardiff"
1
"London"
globalScore
1
windowScore
1
95.65%
Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences
46
glossingSentenceCount
1
matches
0
"felt like the safest place she’d ever k"
11.14%
"Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches
2
per1kWords
1.889
wordCount
1059
matches
0
"not just Evan, but the ghosts of her past"
1
"not Evan, but someone else, someone who’d been part of a different kind of"
100.00%
Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount
0
totalSentences
61
matches
(empty)
87.60%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
15
mean
70.6
std
32.24
cv
0.457
sampleLengths
0
128
1
66
2
133
3
73
4
54
5
64
6
30
7
78
8
47
9
30
10
51
11
54
12
65
13
58
14
128
92.02%
Passive voice overuse
Target: ≤2% passive sentences
passiveCount
2
totalSentences
53
matches
0
"was sprawled"
1
"was slicked"
100.00%
Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount
1
totalVerbs
126
matches
0
"was offering"
49.18%
Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount
1
semicolonCount
1
flaggedSentences
2
totalSentences
61
ratio
0.033
matches
0
"The ancient hunger, the supernatural threat – it all tied back to the world Lucien inhabited, the world she’d tried to escape."
1
"The choice wasn’t just about him; it was about everything she’d fought to build, everything she’d lost, and everything she might lose again."
87.93%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
818
adjectiveStacks
0
stackExamples
(empty)
adverbCount
44
adverbRatio
0.05378973105134474
lyAdverbCount
12
lyAdverbRatio
0.014669926650366748
100.00%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
61
echoCount
0
echoWords
(empty)
100.00%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
61
mean
17.36
std
10.09
cv
0.581
sampleLengths
0
22
1
27
2
27
3
32
4
20
5
14
6
21
7
14
8
17
9
23
10
19
11
2
12
10
13
21
14
15
15
14
16
27
17
2
18
18
19
25
20
14
21
16
22
14
23
27
24
13
25
30
26
14
27
20
28
22
29
8
30
6
31
50
32
22
33
23
34
24
35
7
36
14
37
9
38
4
39
11
40
36
41
5
42
11
43
22
44
9
45
7
46
12
47
53
48
9
49
13
60.66%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
8
diversityRatio
0.4426229508196721
totalSentences
61
uniqueOpeners
27
100.00%
Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount
2
totalSentences
51
matches
0
"Instead, the flat felt suddenly"
1
"Instead, he lifted his cane,"
ratio
0.039
23.92%
Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount
25
totalSentences
51
matches
0
"Her flat was a cramped,"
1
"She’d fled Cardiff, not just"
2
"she murmured, stepping inside, the"
3
"She’d just finished a late"
4
"She’d planned a quiet evening:"
5
"She’d thought she’d left all"
6
"They’d met in the tangled"
7
"He’d been a fixer, impeccably"
8
"He’d helped her, once, but"
9
"She’d thought she’d seen the"
10
"His platinum blond hair was"
11
"He didn’t look like he"
12
"he said, his voice smooth,"
13
"She didn’t move, didn’t speak,"
14
"she finally said, her voice"
15
"He chuckled, a low, dry"
16
"He took a step closer,"
17
"He paused, his amber eye"
18
"she replied, her voice tight"
19
"She forced herself to meet"
ratio
0.49
18.82%
Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount
45
totalSentences
51
matches
0
"The smell of curry andstale"
1
"Her flat was a cramped,"
2
"She’d fled Cardiff, not just"
3
"The tabby cat, Ptolemy, a"
4
"she murmured, stepping inside, the"
5
"She’d just finished a late"
6
"She’d planned a quiet evening:"
7
"She’d thought she’d left all"
8
"The name echoed in her"
9
"They’d met in the tangled"
10
"He’d been a fixer, impeccably"
11
"He’d helped her, once, but"
12
"She’d thought she’d seen the"
13
"His platinum blond hair was"
14
"The scent of expensive cologne,"
15
"He didn’t look like he"
16
"he said, his voice smooth,"
17
"The names rolled off his"
18
"She didn’t move, didn’t speak,"
19
"Ptolemy, sensing a stranger, hissed"
ratio
0.882
98.04%
Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount
1
totalSentences
51
matches
0
"Now, he stood on her"
ratio
0.02
100.00%
Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences
34
technicalSentenceCount
1
matches
0
"But here he was, not Evan, but someone else, someone who’d been part of a different kind of chaos."
0.00%
Useless dialogue additions
Target: ≤5% dialogue tags with trailing filler fragments
totalTags
10
uselessAdditionCount
3
matches
0
"He took, his polished shoes clicking on the worn floorboards"
1
"she replied, her voice tight"
2
"she demanded, the cool mask slipping slightly"
38.89%
Dialogue tag variety (said vs. fancy)
Target: ≤10% fancy dialogue tags
totalTags
5
fancyCount
2
fancyTags
0
"she murmured (murmur)"
1
"she demanded (demand)"
dialogueSentences
18
tagDensity
0.278
leniency
0.556
rawRatio
0.4
effectiveRatio
0.222
78.2127%