Run: 2026-09-23-01M37PRY9NH7EMG824DS1VVR3V

Run Details

Duration:
50.5s
Cost:
$0.017054
Provider:
AionLabs

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="4417" name="Tomás Herrera" aliases="Tommy, Tomás, Herrera" pronouns="he/him" species="Human" storyRole="Support"> <eyeColor>warm brown</eyeColor> <hair>short curly dark brown</hair> <height>5'10"</height> <age>29</age> <skinTone>olive skin</skinTone> <scar>scar running along his left forearm from a knife attack</scar> <medallion>Saint Christopher medallion around his neck</medallion> <birthplace>Seville, Spain</birthplace> <description>A former paramedic who provides off-the-books medical care for the clique.</description> <backstory>Born in Seville, Spain. He moved to London to work for the NHS but lost his license after administering unauthorized treatments to supernatural patients.</backstory> </character> <location id="854" name="The Raven's Nest" aliases="the Nest, Raven, Silas' bar" realm="Earth"> <signColor>distinctive green neon sign above the entrance</signColor> <secretRoom>hidden back room accessible through a bookshelf, used for clandestine meetings</secretRoom> <description>A dimly lit bar in Soho, London. The walls are covered with old maps and black-and-white photographs.</description> </location> <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> </codex> Write a fully fleshed out scene of at least 1000 words in the thriller genre. A detective is pursuing a suspect through rain-soaked city streets at night. The chase takes them from the street level down into an underground market. The detective must decide whether to follow the suspect into unfamiliar and potentially dangerous territory. 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
The rain came down like it had a grudge against London, and Detective Harlow Quinn had been sitting in it for three hours. She'd parked the unmarked Vauxhall two streets up from the green neon sign of the Raven's Nest, close enough to watch the door, far enough that the regulars wouldn't clock her as police. Eighteen years on the job had taught her that much. It hadn't taught her what she was actually looking at. That was the problem with this case. The bar in Soho. The people who drank there. The way witnesses went quiet, then went missing, then turned up across the river with stories that made no sense and no stories at all. Three years ago, DS Morris had knocked on a door much like the one she was watching now. Four days later, Quinn had identified what was left of him by his dental records and a ring she'd bought him for his birthday. Nobody could tell her what happened in between. She'd stopped asking out loud, but she'd never stopped asking. At 1:47 a.m., the door opened and a man stepped out under the green sign. Quinn straightened in her seat. Tomás Herrera. Twenty-nine, olive-skinned, short curly hair plastered down by the rain. Former paramedic, struck off the register for administering unauthorized treatments — the file had been vague about what "unauthorized" meant, and the vagueness itself was interesting. He had a canvas satchel slung across his chest and he was walking fast, head down, like a man who knew exactly where the cameras were. She got out of the car. "Herrera!" Her voice cut through the rain. She held up her warrant card, a pale rectangle in the dark. "Met Police. I'd like a word." He ran. Of course he ran. They always ran, and it always told her something. Herrera sprinted north with his head low, and Quinn followed, boots hammering wet pavement, her breath tearing white plumes out of her. He was quick. He took the first corner hard, shoulder brushing brick, and cut down an alley that smelled of fryer grease and urine. Quinn took the same corner three seconds behind him and felt her feet skid on the slick cobbles — she caught herself on a drainpipe, felt the metal shudder, kept going. Ahead, Herrera hurdled a chain-link fence with an economy that said he'd done this before, and as he landed she heard something small and hard ring off the pavement and skitter into the gutter. He didn't stop for it. He kept running. Quinn vaulted the fence, landed, and swept the gutter without breaking stride — old habit, you took what a runner dropped because it was always something he couldn't afford to leave behind. Her fingers closed on it. Small. Bone-white. Carved with grooves that fit a thumbprint, worn smooth by handling. Some kind of token. She shoved it in her coat pocket and ran on. The chase unspooled north through streets she half-knew. Herrera wasn't running like a man panicking; that was what unsettled her. He ran like a man who knew the route. Left at the shuttered bakery. Right past the kebab shop with its dead neon. Across a road where a night bus hissed past and for two awful seconds she lost him in the spray of its wake — then caught the flash of a Saint Christopher medallion swinging at his throat as he turned his head to check behind him. The patron saint of travelers. She'd have laughed if she'd had the breath for it. Her radio crackled against her shoulder. She ignored it. No backup in Soho inside ten minutes, and Herrera would be gone inside two. He cut through a construction site, ducked under caution tape, and then did something that stopped her cold at the chain-link edge of it: he dropped into a Tube entrance. Not a live station. The iron gates were chained and padlocked, grates bolted across the stairwell, the roundel above the tunnel mouth so faded it was a ghost of itself. An abandoned station. Quinn knew the type — London was riddled with them, dead stations bricked off from the living network. Nobody went down into them. Herrera went down into this one. He'd slipped through a gap where the grating had been bent outward, just wide enough for a lean man with a satchel, and now his footsteps were descending into the dark, quick and certain. Quinn stood at the gap with the rain hammering her shoulders and made herself think. Her watch — the worn leather band gone soft with years of her wrist — read 2:03. She touched the face of it without meaning to, the way she always did when a decision mattered. Down there was unfamiliar ground, no witnesses, no backup, no signal for her radio. Down there was whatever Herrera was so desperate to protect, and whatever had emptied the last three years of her life into a hole she couldn't see the bottom of. Down there, possibly, was the answer to what happened to Morris. She went through the gap. The descent changed everything. The rain cut off above her like a door closing. The stairs were slick with condensation and something else, something faintly oily that she didn't want to name. Her torch was in the car — she'd left it on the passenger seat like an amateur, because she'd expected to have a conversation, not a footrace through half of north London. She went down by feel, one hand on the cold rail, and the darkness was so total that when she saw the light ahead, it hit her like a physical thing. Orange. Flickering. Not electric. She slowed. The stairwell opened onto an old ticket hall, and at the foot of the stairs a figure stood beside a turnstile that had been wrenched out of its housing and repurposed as a gate. The figure was tall, wrapped in a coat too heavy for the season, and its face was lost in the hood. Quinn pressed herself into the shadow of a tiled pillar — decades of soot and a cracked mosaic of some countryside scene that had no business being underground — and watched. Herrera stood at the gate, chest heaving, rain dripping off him. He was talking fast, low. Quinn caught fragments. "—following me. Detective. She's Met, she saw the Nest—" "Token," the gatekeeper said. The voice was wrong. Not male, not female, with a rattle in it like wind through a dead chimney. Herrera's hand went to his pocket and stopped. Quinn saw the exact moment he understood. His fingers searched the satchel, the other pocket, the lining of his jacket. His whole body went still, and then he turned and looked back up the stairs, straight toward the dark where she was hiding, and she watched his face change — not relief, not fear. Calculation. He was deciding what she was. He turned back to the gatekeeper. "I dropped it. Someone took it off the street, it doesn't matter, you know me—" "Token," the gatekeeper said again, unmoving, and now Quinn understood what was in her pocket. The bone-white thing with the groove for a thumb. Not a keepsake. A ticket. Herrera argued. The gatekeeper didn't. Beyond them, through an archway where the ticket barriers used to be, Quinn could see light — warm, many-colored, firelight and lantern-glow — and hear, impossibly, the murmur of a crowd. A market. Underground, at two in the morning, in a station that had been dead for sixty years. The sound carried smells with it: incense, hot metal, something herbal and bitter that made the back of her tongue ache. Her radio hissed static against her shoulder. Useless. Her warrant card was a joke down here. Every instinct trained into her over eighteen years said pull back, mark the location, come back with a full team and daylight. But daylight would find the market gone. She knew it the way she knew the case files were lying to her — this whole world slid out from under procedure like water under a bridge. And Herrera had come out of Silas's bar with a satchel he'd rather run from the police than open, and three years ago Morris had knocked on a door in this world and never come home. She had a token in her pocket that didn't belong to her. Quinn took a breath of the strange, incense-heavy air, felt her heart settle into the steady rhythm she'd learned to force in bad places, and stepped out of the shadow of the pillar. "Evening," she said, and walked toward the gate with her thumb already closed around the bone.
created_at"2026-09-23T17:59:47.785Z" 9/23/2026, 5:59:47 PM
model"aion-labs/aion-3.5-20260923"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency1125
moderation_latency(null)
generation_time50393
tokens_prompt888
tokens_completion3100
native_tokens_prompt811
native_tokens_completion2741
native_tokens_completion_images(null)
native_tokens_reasoning1035
native_tokens_cached811
num_media_prompt(null)
num_input_audio_prompt(null)
num_media_completion0
num_search_results(null)
num_fetches(null)
web_search_engine(null)
origin"https://ncbench.com/"
is_byokfalse
finish_reason"stop"
native_finish_reason"stop"
service_tier(null)
usage0.01705425
router(null)
user_agent"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer(null)
request_id"req-1790186387-MpFNyAe1t0HBwq8gjDJI"
session_id(null)
preset_id(null)
api_type"completions"
id"gen-1790186387-2HvkOvhBwvu2oQQAXbl5"
upstream_id"f5086d87fc0541fc8c650b6b55def92b"
provider_responses
0
endpoint_id"09b1fc56-3888-42bd-b7a0-926851c7f9d6"
id"f5086d87fc0541fc8c650b6b55def92b"
is_byokfalse
latency1125
model_permaslug"aion-labs/aion-3.5-20260923"
provider_name"AionLabs"
status200
total_cost0.01705425
cache_discount0.00182475
upstream_inference_cost0
provider_name"AionLabs"
response_cache_source_id(null)
data_region"global"
workspace_id"97e315e5-d303-487d-83c1-83180e8a13d4"

Evaluation details

Result Evaluator Details Meta Data
0.00% Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags5
adverbTagCount1
adverbTags
0"the gatekeeper said again [again]"
dialogueSentences8
tagDensity0.625
leniency1
rawRatio0.2
effectiveRatio0.2
100.00% AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount1442
totalAiIsmAdverbs0
found(empty)
highlights(empty)
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)
86.13% AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount1442
totalAiIsms4
found
0
word"unsettled"
count1
1
word"footsteps"
count1
2
word"electric"
count1
3
word"mosaic"
count1
highlights
0"unsettled"
1"footsteps"
2"electric"
3"mosaic"
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
narrationSentences110
matches(empty)
100.00% Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount2
hedgeCount1
narrationSentences110
filterMatches
0"think"
1"watch "
hedgeMatches
0"happened to"
100.00% Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences112
gibberishSentences0
adjustedGibberishSentences0
longSentenceCount0
runOnParagraphCount0
giantParagraphCount0
wordSaladCount0
repetitionLoopCount0
controlTokenCount0
repeatedSegmentCount0
maxSentenceWordsSeen45
ratio0
matches(empty)
100.00% Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans0
markdownWords0
totalWords1455
ratio0
matches(empty)
100.00% Missing dialogue indicators (quotation marks)
Target: ≤10% speech attributions without quotation marks
totalAttributions6
unquotedAttributions0
matches(empty)
100.00% Name drop frequency
Target: ≤1.0 per-name mentions per 100 words
totalMentions44
wordCount1428
uniqueNames15
maxNameDensity0.98
worstName"Quinn"
maxWindowNameDensity2
worstWindowName"Quinn"
discoveredNames
London3
Detective2
Harlow1
Quinn14
Vauxhall1
Raven1
Nest1
Soho2
Morris3
Herrera11
Saint1
Christopher1
Tube1
Met1
Silas1
persons
0"Harlow"
1"Quinn"
2"Raven"
3"Morris"
4"Herrera"
5"Saint"
6"Christopher"
7"Silas"
places
0"London"
1"Vauxhall"
2"Soho"
3"Tube"
globalScore1
windowScore1
100.00% Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences74
glossingSentenceCount0
matches(empty)
100.00% "Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches0
per1kWords0
wordCount1455
matches(empty)
100.00% Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount0
totalSentences112
matches(empty)
100.00% Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs35
mean41.57
std31.24
cv0.752
sampleLengths
023
194
260
315
469
56
625
72
835
989
108
1164
12104
1323
1430
1556
1640
1715
1890
195
2095
214
2288
2319
249
2523
2669
2721
2829
2975
3038
3171
3212
3333
3416
92.50% Passive voice overuse
Target: ≤2% passive sentences
passiveCount4
totalSentences110
matches
0"were chained"
1"was riddled"
2"been wrenched"
3"was lost"
0.00% Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount8
totalVerbs235
matches
0"was actually looking"
1"was watching"
2"was walking"
3"were descending"
4"was talking"
5"was hiding"
6"was deciding"
7"were lying"
0.00% Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount15
semicolonCount1
flaggedSentences13
totalSentences112
ratio0.116
matches
0"Former paramedic, struck off the register for administering unauthorized treatments — the file had been vague about what \"unauthorized\" meant, and the vagueness itself was interesting."
1"Quinn took the same corner three seconds behind him and felt her feet skid on the slick cobbles — she caught herself on a drainpipe, felt the metal shudder, kept going."
2"Quinn vaulted the fence, landed, and swept the gutter without breaking stride — old habit, you took what a runner dropped because it was always something he couldn't afford to leave behind."
3"Herrera wasn't running like a man panicking; that was what unsettled her."
4"Across a road where a night bus hissed past and for two awful seconds she lost him in the spray of its wake — then caught the flash of a Saint Christopher medallion swinging at his throat as he turned his head to check behind him."
5"Quinn knew the type — London was riddled with them, dead stations bricked off from the living network."
6"Her watch — the worn leather band gone soft with years of her wrist — read 2:03."
7"Her torch was in the car — she'd left it on the passenger seat like an amateur, because she'd expected to have a conversation, not a footrace through half of north London."
8"Quinn pressed herself into the shadow of a tiled pillar — decades of soot and a cracked mosaic of some countryside scene that had no business being underground — and watched."
9"\"—following me. Detective. She's Met, she saw the Nest—\""
10"His whole body went still, and then he turned and looked back up the stairs, straight toward the dark where she was hiding, and she watched his face change — not relief, not fear."
11"Beyond them, through an archway where the ticket barriers used to be, Quinn could see light — warm, many-colored, firelight and lantern-glow — and hear, impossibly, the murmur of a crowd."
12"She knew it the way she knew the case files were lying to her — this whole world slid out from under procedure like water under a bridge."
100.00% Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount1069
adjectiveStacks0
stackExamples(empty)
adverbCount27
adverbRatio0.025257249766136577
lyAdverbCount6
lyAdverbRatio0.005612722170252572
100.00% Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences112
echoCount0
echoWords(empty)
100.00% Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences112
mean12.99
std10.64
cv0.819
sampleLengths
023
133
210
310
47
54
65
725
818
924
108
1110
1215
135
142
1510
1626
1726
186
197
2012
216
222
234
249
2522
263
2721
2831
2934
305
313
3232
335
341
351
3611
374
3810
398
4012
419
425
439
4446
455
4610
476
483
4914
58.63% Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats11
diversityRatio0.4107142857142857
totalSentences112
uniqueOpeners46
34.36% Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount1
totalSentences97
matches
0"Of course he ran."
ratio0.01
75.67% Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount35
totalSentences97
matches
0"She'd parked the unmarked Vauxhall"
1"It hadn't taught her what"
2"She'd stopped asking out loud,"
3"He had a canvas satchel"
4"She got out of the"
5"Her voice cut through the"
6"She held up her warrant"
7"They always ran, and it"
8"He was quick."
9"He took the first corner"
10"He didn't stop for it."
11"He kept running."
12"Her fingers closed on it."
13"She shoved it in her"
14"He ran like a man"
15"She'd have laughed if she'd"
16"Her radio crackled against her"
17"She ignored it."
18"He cut through a construction"
19"He'd slipped through a gap"
ratio0.361
78.56% Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount74
totalSentences97
matches
0"The rain came down like"
1"She'd parked the unmarked Vauxhall"
2"It hadn't taught her what"
3"That was the problem with"
4"The bar in Soho."
5"The people who drank there."
6"The way witnesses went quiet,"
7"Nobody could tell her what"
8"She'd stopped asking out loud,"
9"Quinn straightened in her seat."
10"He had a canvas satchel"
11"She got out of the"
12"Her voice cut through the"
13"She held up her warrant"
14"They always ran, and it"
15"Herrera sprinted north with his"
16"He was quick."
17"He took the first corner"
18"Quinn took the same corner"
19"He didn't stop for it."
ratio0.763
0.00% Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount0
totalSentences97
matches(empty)
ratio0
61.99% Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences53
technicalSentenceCount6
matches
0"The way witnesses went quiet, then went missing, then turned up across the river with stories that made no sense and no stories at all."
1"Herrera sprinted north with his head low, and Quinn followed, boots hammering wet pavement, her breath tearing white plumes out of her."
2"Ahead, Herrera hurdled a chain-link fence with an economy that said he'd done this before, and as he landed she heard something small and hard ring off the pave…"
3"He cut through a construction site, ducked under caution tape, and then did something that stopped her cold at the chain-link edge of it: he dropped into a Tube…"
4"The stairwell opened onto an old ticket hall, and at the foot of the stairs a figure stood beside a turnstile that had been wrenched out of its housing and repu…"
5"Quinn pressed herself into the shadow of a tiled pillar — decades of soot and a cracked mosaic of some countryside scene that had no business being underground …"
100.00% Useless dialogue additions
Target: ≤5% dialogue tags with trailing filler fragments
totalTags5
uselessAdditionCount0
matches(empty)
100.00% Dialogue tag variety (said vs. fancy)
Target: ≤10% fancy dialogue tags
totalTags3
fancyCount0
fancyTags(empty)
dialogueSentences8
tagDensity0.375
leniency0.75
rawRatio0
effectiveRatio0
79.5950%