Run: 2026-09-24-01M39WAY3X7E7R9VGQQMH6WQ79

Run Details

Duration:
46.0s
Cost:
$0.003656
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 a verdict, and Detective Harlow Quinn ran through it with her collar up and her breath burning in her chest. She'd made him outside the Raven's Nest. That was the mistake — hers or his, she hadn't decided yet. She'd been watching the bar for six nights, nursing the same lager for three hours at a stretch, watching the green neon sign flicker above the door like a sick heartbeat. Tonight a man in a waxed coat had come out of the alley beside it carrying a leather satchel that rode too heavy on his shoulder for the clothes inside it. He'd handed it to nobody. He'd simply stood beneath the sign, tilted his face to the rain, and looked directly at her window table through the glass. Then he ran. Now he was half a street ahead, his coat slick and black under the sodium lamps, and Soho blurred past her in smears of neon and puddle-light. Her left wrist ached where the old leather watch bit into it — three years of wear, three years since she'd last taken it off, three years since Morris — and she ran with her thumb pressed against the crystal like it was a talisman of her own. "Witness! Stop — police!" Her voice cracked against the rain and died there. Nobody in Soho stopped for shouts at midnight. Nobody in Soho stopped for anything. He cut left off Wardour Street, through a service alley where the rain drummed on fire escapes and the bins exhaled the smell of rot. She took the corner too tight, her shoulder kissing wet brick, and gained three metres. He was older than her, heavier through the middle — she should be closing. Instead he moved like a man who knew every drain and dead end, like he'd rehearsed this run the way other people rehearsed wedding toasts. The satchel bounced against his hip. Whatever was in it, it wasn't papers. He crossed Charing Cross Road against traffic — a cab's horn tore the night open — and she followed, her boots skidding on the painted crossing, her heart hammering out a rhythm she recognized from the worst night of her career. Not tonight. Not again. She had buried one partner with questions left in the ground beside him, and she would not bury this case the same way. North. He was heading north, threading through Bloomsbury's dark squares, and the geography of it began to make an ugly kind of sense. Camden. He was running for Camden. By the time the streetlights thinned out around Camden High Street, the rain had settled into a steady hiss and her lungs had filed their complaint and been overruled. He turned off the main road into a lane she didn't have a name for, and there — she nearly lost him — he dropped down a flight of iron steps into the dark and was gone. She stopped at the top of the steps, rain streaming off her nose. Below her yawned the mouth of an old Tube station, one of those that had closed before she was born, its tiles grimed grey-green, its name half-legible in flaking enamel. A padlocked gate hung open on its chain, the lock's shackle neatly cut, the cut bright and new. The air rising out of the ground was warm, which was wrong, because the air in a dead station should have been cold and mineral and dead itself. This air moved. It smelled, ridiculously, of candle wax and cloves. Her torch found the platform below, littered with decades of drift, and one set of footprints cutting through it, fresh, heading into the black of the running tunnel. Quinn stood at the top of the steps and made herself do the arithmetic. No backup — she hadn't called any, because she hadn't wanted to explain what she was doing watching a Soho bar for six nights on a hunch stitched from a dead man's file. No warrant. No vest. A suspect with a heavy satchel running into a sealed tunnel like he had an appointment. Every professional instinct she owned said wait. Seal it, get Transport Police down here, flood the place with torchlight and numbers at dawn. But dawn was eleven hours away, and the man in the waxed coat had looked at her through the window of the Raven's Nest like he knew exactly which file her face belonged to. Like Morris's file. Like he'd been sent to test whether anyone was still pulling that thread. She went down the steps. The platform surrendered her footfalls to silence. Her torch beam swept paper cups, a rusted trolley wheel, a pigeon's skeleton folded like a dropped glove. At the tunnel mouth, the footprints ended at a patch of wall that should have been curved tile and was, instead, flat. Her beam crawled across it twice before her brain accepted the information. The wall was new. Brick, mortared, flush — a wall that had no business in a tunnel that had been abandoned for half a century, and no business standing in a tunnel that had been abandoned tonight. Except it wasn't standing. As she watched, the mortar lines glowed faintly, like heat through skin, and a seam opened down the middle. Through the gap she caught a sliver of light — amber, crowded, alive — and heard, impossibly, the murmur of a hundred conversations. On the ground before the wall lay a small white object. Bone, she thought, carved and bored like a token from some grotesque arcade. The man had dropped it, or spent it, or left it. Her hand closed around it before she could argue with herself. It was warm. It had a pulse in it, faint and slow, like a sleeping animal. The seam was closing. Quinn did the thing she'd spent eighteen years training herself not to do and made the decision in under a second: she put her shoulder to the gap and slid through sideways, into the light, into the noise, before the wall sealed behind her with a sound like a held breath released. She stood in a market that did not exist. An abandoned platform stretched away under vaulted ceiling, and it was lined — stalls, trestles, awnings, lanterns — with vendors selling things her vocabulary refused. Bottles that glowed from the inside. Cages of moths with human eyes. A woman weighing smoke on brass scales. The crowd moved through it all wearing coats from every decade at once, and not one of them looked at the newcomer the way strangers look at strangers. They looked at her the way shopkeepers look at weather: briefly, and with calculation. Her torch was off. She didn't remember switching it off. She pocketed the bone token and moved down the aisle, scanning for the waxed coat, and instead found a face she knew. Tomás Herrera — she'd sat across an interview room table from him two years ago, before they'd taken his license, before the NHS washed its hands of him. He was kneeling over a kid on a camp cot behind a stall of dried herbs, his sleeve rolled up off the scar on his left forearm, a Saint Christopher medallion swinging free as he worked. The kid's arm was wrong in a way that made her stomach drop, and Herrera was fixing it with his hands and a thread of light that came from nowhere. He looked up. His warm brown eyes went flat when he found hers. "Detective Quinn." He didn't stop working. "You're a long way from the paperwork." "I'm looking for a man in a waxed coat. He came through that wall two minutes ago." "So did a lot of people. Half of them are lying about why." He tied off his work and stood, wiping his hands. "You shouldn't have come in here. That token you're carrying — it knows you now. It doesn't let you leave the same way you came in." "Meaning what?" "Meaning," Herrera said, and glanced past her shoulder at something in the crowd, "you'd better decide quickly whether you're a visitor or a resident." Quinn turned. The wall she'd come through was gone. In its place stood the man in the waxed coat, satchel open in his hands, watching her with the patience of someone whose appointment had just arrived. "Detective Quinn," he said. "I was told you might come. I have a message about DS Morris." Her watch bit into her wrist. She did not run.
created_at"2026-09-24T14:15:29.162Z" 9/24/2026, 2:15:29 PM
model"aion-labs/aion-3.5-mini-20260923"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency2636
moderation_latency(null)
generation_time45956
tokens_prompt888
tokens_completion2875
native_tokens_prompt811
native_tokens_completion2507
native_tokens_completion_images(null)
native_tokens_reasoning884
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)
usage0.00365578
router(null)
user_agent"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer(null)
request_id"req-1790259329-4FpXfTitldUR7xOPDt4f"
session_id(null)
preset_id(null)
api_type"completions"
id"gen-1790259329-J6hXfrRIMLmnZCzgfcwm"
upstream_id"c7662d37eda44bc68cb1831a77dfd863"
provider_responses
0
endpoint_id"60ac3bdc-90ec-4769-9193-2c00b4d79562"
id"c7662d37eda44bc68cb1831a77dfd863"
is_byokfalse
latency2635
model_permaslug"aion-labs/aion-3.5-mini-20260923"
provider_name"AionLabs"
status200
total_cost0.00365578
cache_discount0.00042172
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
100.00% Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags5
adverbTagCount0
adverbTags(empty)
dialogueSentences10
tagDensity0.5
leniency1
rawRatio0
effectiveRatio0
96.46% AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount1411
totalAiIsmAdverbs1
found
0
adverb"quickly"
count1
highlights
0"quickly"
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)
75.19% AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount1411
totalAiIsms7
found
0
word"flicker"
count1
1
word"crystal"
count1
2
word"streaming"
count1
3
word"silence"
count1
4
word"pulse"
count1
5
word"scanning"
count1
6
word"stomach"
count1
highlights
0"flicker"
1"crystal"
2"streaming"
3"silence"
4"pulse"
5"scanning"
6"stomach"
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
narrationSentences88
matches(empty)
61.69% Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount4
hedgeCount1
narrationSentences88
filterMatches
0"watch"
1"look"
hedgeMatches
0"began to"
100.00% Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences93
gibberishSentences0
adjustedGibberishSentences0
longSentenceCount0
runOnParagraphCount0
giantParagraphCount0
wordSaladCount0
repetitionLoopCount0
controlTokenCount0
repeatedSegmentCount0
maxSentenceWordsSeen52
ratio0
matches(empty)
100.00% Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans0
markdownWords0
totalWords1428
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
totalMentions35
wordCount1330
uniqueNames21
maxNameDensity0.3
worstName"Quinn"
maxWindowNameDensity1.5
worstWindowName"Soho"
discoveredNames
Detective1
Harlow1
Quinn4
Raven2
Nest2
Soho4
Morris2
Wardour1
Street2
Charing1
Cross1
Road1
Bloomsbury1
Camden3
High1
Tube1
Transport1
Police1
Herrera3
Saint1
Christopher1
persons
0"Harlow"
1"Quinn"
2"Raven"
3"Morris"
4"Herrera"
5"Saint"
6"Christopher"
places
0"Soho"
1"Wardour"
2"Street"
3"Charing"
4"Cross"
5"Road"
6"Bloomsbury"
7"Camden"
8"High"
globalScore1
windowScore1
100.00% Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences66
glossingSentenceCount0
matches(empty)
100.00% "Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches0
per1kWords0
wordCount1428
matches(empty)
100.00% Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount0
totalSentences93
matches(empty)
100.00% Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs35
mean40.8
std31.82
cv0.78
sampleLengths
025
1108
23
375
427
579
613
768
829
966
1013
1187
1228
1367
1423
1550
165
1796
1846
1935
2027
2156
229
2386
2410
25116
2613
2713
2817
2949
302
3124
3236
3317
3410
77.35% Passive voice overuse
Target: ≤2% passive sentences
passiveCount7
totalSentences88
matches
0"been overruled"
1"was gone"
2"been sent"
3"been curved"
4"been abandoned"
5"been abandoned"
6"was lined "
7"was gone"
0.00% Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount7
totalVerbs222
matches
0"was heading"
1"was running"
2"was doing"
3"was still pulling"
4"wasn't standing"
5"was kneeling"
6"was fixing"
0.00% Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount15
semicolonCount0
flaggedSentences10
totalSentences93
ratio0.108
matches
0"That was the mistake — hers or his, she hadn't decided yet."
1"Her left wrist ached where the old leather watch bit into it — three years of wear, three years since she'd last taken it off, three years since Morris — and she ran with her thumb pressed against the crystal like it was a talisman of her own."
2"He was older than her, heavier through the middle — she should be closing."
3"He crossed Charing Cross Road against traffic — a cab's horn tore the night open — and she followed, her boots skidding on the painted crossing, her heart hammering out a rhythm she recognized from the worst night of her career."
4"He turned off the main road into a lane she didn't have a name for, and there — she nearly lost him — he dropped down a flight of iron steps into the dark and was gone."
5"No backup — she hadn't called any, because she hadn't wanted to explain what she was doing watching a Soho bar for six nights on a hunch stitched from a dead man's file."
6"Brick, mortared, flush — a wall that had no business in a tunnel that had been abandoned for half a century, and no business standing in a tunnel that had been abandoned tonight."
7"Through the gap she caught a sliver of light — amber, crowded, alive — and heard, impossibly, the murmur of a hundred conversations."
8"An abandoned platform stretched away under vaulted ceiling, and it was lined — stalls, trestles, awnings, lanterns — with vendors selling things her vocabulary refused."
9"Tomás Herrera — she'd sat across an interview room table from him two years ago, before they'd taken his license, before the NHS washed its hands of him."
100.00% Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount1318
adjectiveStacks0
stackExamples(empty)
adverbCount28
adverbRatio0.021244309559939303
lyAdverbCount10
lyAdverbRatio0.007587253414264037
100.00% Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences93
echoCount0
echoWords(empty)
100.00% Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences93
mean15.35
std11.65
cv0.759
sampleLengths
025
17
212
331
431
55
622
73
827
948
1013
118
126
1325
1415
1514
1625
176
187
1941
202
212
2223
231
2422
251
265
2729
2837
2913
3030
3118
3228
333
348
3528
3614
3733
382
392
4016
417
4216
4334
443
4513
465
477
4818
4922
70.97% Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats9
diversityRatio0.4838709677419355
totalSentences93
uniqueOpeners45
83.33% Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount2
totalSentences80
matches
0"Then he ran."
1"Instead he moved like a"
ratio0.025
45.00% Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount35
totalSentences80
matches
0"She'd made him outside the"
1"She'd been watching the bar"
2"He'd handed it to nobody."
3"He'd simply stood beneath the"
4"Her left wrist ached where"
5"Her voice cracked against the"
6"He cut left off Wardour"
7"She took the corner too"
8"He was older than her,"
9"He crossed Charing Cross Road"
10"She had buried one partner"
11"He was heading north, threading"
12"He was running for Camden."
13"He turned off the main"
14"She stopped at the top"
15"It smelled, ridiculously, of candle"
16"Her torch found the platform"
17"She went down the steps."
18"Her torch beam swept paper"
19"Her beam crawled across it"
ratio0.438
72.50% Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount62
totalSentences80
matches
0"The rain came down like"
1"She'd made him outside the"
2"That was the mistake —"
3"She'd been watching the bar"
4"Tonight a man in a"
5"He'd handed it to nobody."
6"He'd simply stood beneath the"
7"Her left wrist ached where"
8"Her voice cracked against the"
9"Nobody in Soho stopped for"
10"Nobody in Soho stopped for"
11"He cut left off Wardour"
12"She took the corner too"
13"He was older than her,"
14"The satchel bounced against his"
15"Whatever was in it, it"
16"He crossed Charing Cross Road"
17"She had buried one partner"
18"He was heading north, threading"
19"He was running for Camden."
ratio0.775
100.00% Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount2
totalSentences80
matches
0"Now he was half a"
1"By the time the streetlights"
ratio0.025
98.21% Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences48
technicalSentenceCount3
matches
0"Tonight a man in a waxed coat had come out of the alley beside it carrying a leather satchel that rode too heavy on his shoulder for the clothes inside it."
1"Instead he moved like a man who knew every drain and dead end, like he'd rehearsed this run the way other people rehearsed wedding toasts."
2"He crossed Charing Cross Road against traffic — a cab's horn tore the night open — and she followed, her boots skidding on the painted crossing, her heart hamme…"
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
totalTags2
fancyCount0
fancyTags(empty)
dialogueSentences10
tagDensity0.2
leniency0.4
rawRatio0
effectiveRatio0
86.0236%