NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-17-01KHMPH17X9JZ7G7M366P16WR9
Run Details
Model:
Llama 3.1 70B
Test:
Bad Writing Habits
Scenario:
Mystery: examining a crime scene
Duration:
47.8s
Cost:
$0.000713
Provider:
DeepInfra
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
Messages
Meta
AI
Detective Harlow Quinn stepped off the dusty platform and onto the worn concrete floor of the abandoned Tube station.
The air was heavy with
the scent of damp earth and decay, a morbid reminder that this place had
been left
to rot. A faint hum of generators
echoed
through the cavernous space, casting flickering shadows on the walls as a team of forensic technicians worked to process the crime scene. Quinn's eyes scanned the area, her sharp jaw set in a determined line. The victim, a young woman with piercing green eyes, lay on the ground, her curly red hair matted with blood. Quinn's gaze lingered on the woman's face, a
pang
of recognition sparking in her chest. Eva Kowalski, a name that had surfaced in several of Quinn's recent investigations. The research assistant from the British Museum's restricted archives had been a peripheral figure, but her connection to the mysterious clique that had been the focus of Quinn's attention for months was undeniable. Quinn's colleague, Detective Inspector Rachel
Patel
, approached her, a look of grim concern
etched
on her face. "Harlow, we've got a mess here. The victim's ID says she's Eva Kowalski, but...I don't know, it just doesn't add up." Quinn raised an eyebrow, her eyes never leaving the body. "What do you mean?"
Patel
hesitated. "Well, the wound patterns look like a ritualistic killing, but there's no sign of forced entry or struggle. It's as if the killer just...appeared. And the strange symbols
etched
into the walls...I've never seen anything like it." Quinn's gaze shifted to the symbols, her mind racing with possibilities. She
knew that Patel
, a seasoned detective with a no-nonsense approach, wouldn't jump to conclusions without evidence. But Quinn's instincts told her that something was off. She took a step closer to the body, her eyes scanning the surrounding area. A small, worn leather satchel lay nearby, its contents spilled across the floor. Quinn spotted a book with yellowed pages, a torn piece of paper with cryptic notes, and a small, brass compass with a verdigris patina. The Veil Compass, an item she had only read about in whispered rumors. Its presence here was a game-changer. "Eva Kowalski was more than just a research assistant," Quinn said, her voice low and even. "She was an occult researcher, and this compass suggests she was involved in something much bigger than we initially thought."
Patel
's expression turned skeptical. "You're not buying into the whole supernatural angle, are you, Harlow? We need concrete evidence, not fantasy." Quinn's sharp jawline tightened. She
knew that Patel
didn't understand the depths of Quinn's investigation, the threads that connected the clique to the mysterious energy rifts that had been popping up around the city. But Quinn had seen the unexplainable, the shadows that lurked just beyond the edge of reality. She
knew that this
case was more than just a simple murder. "I'm telling you, Rachel, this compass is the key. It's a tool for
navigating
the supernatural, and Eva Kowalski's presence here, in this abandoned station...it's all connected to the Veil Market."
Patel
raised an eyebrow. "The Veil Market? You mean that mythical underground black market?" Quinn nodded. "I do. And I think Eva Kowalski was trying to uncover something, something big. This symbol," she pointed to the
etching
on the wall, "is a calling card, a signature. Our killer is trying to send a message, but to whom?" As Quinn continued to examine the crime scene, she noticed a small, almost imperceptible mark on the wall, near the symbol. A tiny, bone token, the entry requirement for the Veil Market,
was embedded
in the concrete. Quinn's eyes met
Patel
's, a spark of understanding passing between them. "We need to dig deeper," Quinn said, her voice firm. "We need to find out what Eva Kowalski was researching, and what she uncovered that got her killed."
Patel
nodded, a determined look on her face. "I'll start going through Eva's research, see if I can find any connections to the Veil Market. You...?" Quinn's gaze drifted back to the Veil Compass, her mind racing with possibilities. "I'll follow the symbol, see where it leads. This case just got a lot more interesting."
created_at
"2026-02-17T02:23:51.672136+00:00"
2/17/2026, 2:23:51 AM
model
"meta-llama/llama-3.1-70b-instruct"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
431
moderation_latency
(null)
generation_time
47677
tokens_prompt
968
tokens_completion
1080
native_tokens_prompt
848
native_tokens_completion
935
native_tokens_completion_images
(null)
native_tokens_reasoning
0
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
0.0007132
router
(null)
provider_responses
0
id
"cmpl-64b996600680451398f11f6d24eb3376"
status
200
is_byok
false
latency
65
endpoint_id
"59c87462-40b6-4231-91cf-6d0f8f25e8b9"
provider_name
"DeepInfra"
model_permaslug
"meta-llama/llama-3.1-70b-instruct"
api_type
"completions"
id
"gen-1771294983-8ot3n0xIDURxOaMsXCJo"
upstream_id
"cmpl-64b996600680451398f11f6d24eb3376"
total_cost
0.0007132
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
3
adverbTagCount
0
adverbTags
(empty)
100.0000%
AI-ism adverb frequency
Target: <1% AI-ism adverbs (57 tracked)
wordCount
702
totalAiIsmAdverbs
0
found
(empty)
highlights
(empty)
80.0000%
AI-ism character names
Target: 0 AI-default names (17 tracked, −20% each)
codexExemptions
(empty)
found
0
"Patel"
100.0000%
AI-ism location names
Target: 0 AI-default location names (33 tracked, −20% each)
codexExemptions
(empty)
found
(empty)
14.5299%
AI-ism word frequency
Target: <1% AI-ism words (140 tracked)
wordCount
702
totalAiIsms
6
found
0
word
"pang"
count
1
1
word
"echoed"
count
1
2
word
"etched"
count
2
3
word
"etching"
count
1
4
word
"navigating"
count
1
highlights
0
"pang"
1
"echoed"
2
"etched"
3
"etching"
4
"navigating"
100.0000%
Cliche density
Target: ≤1 cliche(s) per 800-word window
totalCliches
1
maxInWindow
1
found
0
label
"air was thick with"
count
1
highlights
0
"The air was heavy with"
100.0000%
Emotion telling (show vs. tell)
Target: ≤3% sentences with emotion telling
emotionTells
0
narrationSentences
34
matches
(empty)
100.0000%
Filter word density
Target: ≤12% sentences with filter/hedge words
filterCount
0
hedgeCount
0
narrationSentences
34
filterMatches
(empty)
hedgeMatches
(empty)
100.0000%
Overuse of "that" (subordinate clause padding)
Target: ≤10% sentences with "that" clauses
thatCount
3
totalSentences
44
matches
0
"knew that Patel"
1
"knew that Patel"
2
"knew that this"
95.0634%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
17
mean
41.18
std
19.88
cv
0.483
sampleLengths
0
83
1
81
2
38
3
14
4
39
5
51
6
56
7
36
8
21
9
62
10
31
11
14
12
43
13
48
14
28
15
26
16
29
94.1176%
Passive voice overuse
Target: ≤5% passive sentences
passiveCount
2
totalSentences
34
matches
0
"been left"
1
"was embedded"
100.0000%
Past progressive (was/were + -ing) overuse
Target: ≤10% past progressive verbs
pastProgressiveCount
0
totalVerbs
62
matches
(empty)
100.0000%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
463
adjectiveStacks
0
stackExamples
(empty)
adverbCount
6
adverbRatio
0.012958963282937365
lyAdverbCount
2
lyAdverbRatio
0.004319654427645789
100.0000%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
44
echoCount
0
echoWords
(empty)
100.0000%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
44
mean
15.91
std
8.97
cv
0.564
sampleLengths
0
19
1
23
2
28
3
13
4
20
5
15
6
13
7
33
8
17
9
21
10
10
11
4
12
2
13
37
14
11
15
17
16
9
17
14
18
13
19
24
20
13
21
6
22
16
23
20
24
4
25
17
26
4
27
30
28
16
29
12
30
31
31
4
32
10
33
2
34
41
35
21
36
16
37
11
38
10
39
18
40
8
41
18
42
13
43
16
58.3333%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
1
diversityRatio
0.36363636363636365
totalSentences
44
uniqueOpeners
16
91.6667%
Dialogue tag variety (said vs. fancy)
Target: ≤30% fancy dialogue tags
totalTags
3
fancyCount
1
fancyTags
0
"she pointed to the etching on the wall (point)"
90.2183%