Run: 2026-02-17-01KHMMEWHCGY0N34XBABP9YWS5

Run Details

Model:
Hermes 3 70B
Test:
Bad Writing Habits
Scenario:
Mystery: examining a crime scene
Duration:
17.8s
Cost:
$0.000488
Provider:
DeepInfra

Evaluation summary

Score Evaluator
Details Link
100.00% Adverbs in dialogue tags (details)
74.96% AI-ism adverb frequency (details)
100.00% AI-ism character names (details)
100.00% AI-ism location names (details)
74.96% AI-ism word frequency (details)
100.00% Cliché density (details)
100.00% Emotion telling (show vs. tell) (details)
37.04% Filter word density (details)
100.00% Gibberish response detection (details)
100.00% Markdown formatting overuse (details)
100.00% Missing dialogue indicators (quotation marks) (details)
40.78% Name drop frequency (details)
50.00% Narrator intent-glossing (details)
100.00% "Not X but Y" pattern overuse (details)
71.43% Overuse of "that" (subordinate clause padding) (details)
47.08% Paragraph length variance (details)
100.00% Passive voice overuse (details)
0.00% Past progressive (was/were + -ing) overuse (details)
61.22% Em-dash & semicolon overuse (details)
94.00% Purple prose (modifier overload) (details)
100.00% Repeated phrase echo (details)
95.09% Sentence length variance (details)
78.10% Sentence opener variety (details)
0.00% Adverb-first sentence starts (details)
100.00% Pronoun-first sentence starts (details)
15.56% Subject-first sentence starts (details)
0.00% Subordinate conjunction sentence starts (details)
100.00% Technical jargon density (details)
0.00% Useless dialogue additions (details)
16.67% Dialogue tag variety (said vs. fancy) (details)
68.5624% 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
Detective Harlow Quinn ducked under the yellow tape, her worn leather watch catching the light as she entered the abandoned Tube station. The air was thick with the musty smell of abandonment, punctuated by the coppery tang of fresh blood. She surveyed the scene with a critical eye, her sharp jaw tightening as she took in the details. The body of a young man lay sprawled across the grimy platform, his pale face eerily illuminated by the flickering overhead lights. A swirl of dark crimson pooled beneath him, seeping into the cracks between the tiles. The murder weapon, a simple iron dagger, sat on the ground nearby, its blade cleaned of any incriminating evidence. "Eva Kowalski, what brings you here?" Harlow asked, not bothering to hide the irritation in her voice as she spotted her former informant. Eva, a small, freckle-faced woman with curly red hair, tugged nervously at a strand behind her left ear. "I found something," Eva said, her green eyes wide behind her round glasses. She produced a small brass compass from her worn leather satchel, its face etched with unusual markings. "This is a Veil Compass, used to locate supernatural rifts or portals." Harlow frowned, taking the compass and examining it closely. The needle twitched erratically, pointing at the far end of the platform. "This is nonsense, Eva. You know I don't believe in that supernatural crap." "I know, Harlow, but look around," Eva pleaded, gesturing to the odd symbols scrawled on the walls in what appeared to be a mixture of blood and soot. "This isn't your typical murder scene." Harlow grudgingly acknowledged Eva's point, turning her attention back to the body. Something wasn't adding up, but she couldn't quite put her finger on it. She paced the length of the platform, scrutinizing the scuffs on the ground and the way the blood had spattered across the walls. It was almost as if the killer had been trying to create a specific pattern. "I think this might be connected to a case I've been working on," Eva admitted, her voice barely above a whisper. "The Veil Market whispers about a group of humans dabbling in dangerous occult practices. They call themselves 'the Clique.'" "You've been spending too much time studying the occult, Eva," Harlow said, but her tone lacked conviction. She knew Eva's research had led to breakthroughs in several of her cases, despite her reluctance to admit it. Eva pointed to a symbol on the ground, a circular pattern with several intersecting lines. "This is a sigil, Harlow. It's meant to open a portal or draw something from the other side." "Fine," Harlow said, sighing heavily. "Let's say you're right. What do you think this Clique wanted with our victim?" "I don't know," Eva said, biting her lip. "But I think we need to find out more about the Veil Market. If there's a chance this murder is connected to the occult, we need to tread carefully." Harlow nodded, pocketing the Veil Compass. "Okay, Eva. You work your contacts, see if you can dig up any more information on the Veil Market or this damned Clique. I'll look into our victim, see if I can find any connections." As they parted ways, Harlow couldn't shake the feeling that Eva was right — that they were delving into a world far beyond their understanding. But if there was one thing she'd learned in her 18 years on the force, it was that sometimes, the truth was stranger than fiction. And something was telling her that this case was going to be her strangest one yet.
created_at"2026-02-17T01:47:14.212133+00:00" 2/17/2026, 1:47:14 AM
model"nousresearch/hermes-3-llama-3.1-70b"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency162
moderation_latency(null)
generation_time17730
tokens_prompt945
tokens_completion894
native_tokens_prompt846
native_tokens_completion781
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"
usage0.0004881
router(null)
provider_responses
0
id"cmpl-af574c8d24944b6791a439d73afdf1e1"
status200
is_byokfalse
latency74
endpoint_id"2151651b-f4e9-4db7-b20d-8b91fb524bf8"
provider_name"DeepInfra"
model_permaslug"nousresearch/hermes-3-llama-3.1-70b"
api_type"completions"
id"gen-1771292815-Tu22a1BPYUF5IfuPU4B7"
upstream_id"cmpl-af574c8d24944b6791a439d73afdf1e1"
total_cost0.0004881
cache_discount(null)
upstream_inference_cost0
provider_name"DeepInfra"

Evaluation details

Result Evaluator Details Meta Data
100.00% Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags7
adverbTagCount0
adverbTags(empty)
dialogueSentences15
tagDensity0.467
leniency0.933
rawRatio0
effectiveRatio0
74.96% AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount599
totalAiIsmAdverbs3
found
0
adverb"nervously"
count1
1
adverb"barely above a whisper"
count1
2
adverb"carefully"
count1
highlights
0"nervously"
1"barely above a whisper"
2"carefully"
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)
74.96% AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount599
totalAiIsms3
found
0
word"etched"
count1
1
word"whisper"
count1
2
word"delving"
count1
highlights
0"etched"
1"whisper"
2"delving"
100.00% Cliché density
Target: ≤1 cliche(s) per 800-word window
totalCliches1
maxInWindow1
found
0
label"air was thick with"
count1
highlights
0"The air was thick with"
100.00% Emotion telling (show vs. tell)
Target: ≤3% sentences with emotion telling
emotionTells0
narrationSentences27
matches(empty)
37.04% Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount1
hedgeCount1
narrationSentences27
filterMatches
0"watch"
hedgeMatches
0"appeared to"
100.00% Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences35
gibberishSentences0
adjustedGibberishSentences0
longSentenceCount0
runOnParagraphCount0
giantParagraphCount0
wordSaladCount0
repetitionLoopCount0
controlTokenCount0
maxSentenceWordsSeen35
ratio0
matches(empty)
100.00% Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans0
markdownWords0
totalWords600
ratio0
matches(empty)
100.00% Missing dialogue indicators (quotation marks)
Target: ≤10% speech attributions without quotation marks
totalAttributions10
unquotedAttributions0
matches(empty)
40.78% Name drop frequency
Target: ≤1.0 per-name mentions per 100 words
totalMentions21
wordCount412
uniqueNames6
maxNameDensity2.18
worstName"Eva"
maxWindowNameDensity3.5
worstWindowName"Eva"
discoveredNames
Harlow8
Quinn1
Tube1
Eva9
Veil1
Compass1
persons
0"Harlow"
1"Quinn"
2"Eva"
places(empty)
globalScore0.408
windowScore0.5
50.00% Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences25
glossingSentenceCount1
matches
0"quite put her finger on it"
100.00% "Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches0
per1kWords0
wordCount600
matches(empty)
71.43% Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount1
totalSentences35
matches
0"right — that they"
47.08% Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs15
mean40
std12.59
cv0.315
sampleLengths
058
156
241
342
434
534
663
740
836
933
1019
1137
1241
1350
1416
100.00% Passive voice overuse
Target: ≤2% passive sentences
passiveCount0
totalSentences27
matches(empty)
0.00% Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount3
totalVerbs67
matches
0"wasn't adding"
1"were delving"
2"was telling"
61.22% Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount1
semicolonCount0
flaggedSentences1
totalSentences35
ratio0.029
matches
0"As they parted ways, Harlow couldn't shake the feeling that Eva was right — that they were delving into a world far beyond their understanding."
94.00% Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount412
adjectiveStacks1
stackExamples
0"small, freckle-faced woman"
adverbCount10
adverbRatio0.024271844660194174
lyAdverbCount8
lyAdverbRatio0.019417475728155338
100.00% Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences35
echoCount0
echoWords(empty)
95.09% Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences35
mean17.14
std6.65
cv0.388
sampleLengths
022
118
218
322
415
519
623
718
813
917
1012
119
1212
1313
1428
156
1612
1713
1823
1915
2021
2119
2217
2319
2415
2518
265
2714
288
2929
306
3135
3225
3325
3416
78.10% Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats1
diversityRatio0.4857142857142857
totalSentences35
uniqueOpeners17
0.00% Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount0
totalSentences27
matches(empty)
ratio0
100.00% Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount5
totalSentences27
matches
0"She surveyed the scene with"
1"She produced a small brass"
2"She paced the length of"
3"It was almost as if"
4"She knew Eva's research had"
ratio0.185
15.56% Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount24
totalSentences27
matches
0"Detective Harlow Quinn ducked under"
1"The air was thick with"
2"She surveyed the scene with"
3"The body of a young"
4"A swirl of dark crimson"
5"The murder weapon, a simple"
6"Harlow asked, not bothering to"
7"Eva, a small, freckle-faced woman"
8"Eva said, her green eyes"
9"She produced a small brass"
10"Harlow frowned, taking the compass"
11"The needle twitched erratically, pointing"
12"Eva pleaded, gesturing to the"
13"Harlow grudgingly acknowledged Eva's point,"
14"Something wasn't adding up, but"
15"She paced the length of"
16"It was almost as if"
17"Eva admitted, her voice barely"
18"Harlow said, but her tone"
19"She knew Eva's research had"
ratio0.889
0.00% Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount0
totalSentences27
matches(empty)
ratio0
100.00% Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences18
technicalSentenceCount0
matches(empty)
0.00% Useless dialogue additions
Target: ≤5% dialogue tags with trailing filler fragments
totalTags7
uselessAdditionCount3
matches
0"Eva said, her green eyes wide behind her round glasses"
1"Eva admitted, her voice barely above a whisper"
2"Harlow said, but her tone lacked conviction"
16.67% Dialogue tag variety (said vs. fancy)
Target: ≤10% fancy dialogue tags
totalTags7
fancyCount2
fancyTags
0"Eva pleaded (plead)"
1"Eva admitted (admit)"
dialogueSentences15
tagDensity0.467
leniency0.933
rawRatio0.286
effectiveRatio0.267
68.5624%