NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-27-01KQ73S3VPCYRV4KN9WZK5NEJ9
Run Details
Model:
Qwen 3.6 27B
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
2m 2s
Cost:
$0.029083
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
98.71%
Matches word count
(details)
94.38%
Dialogue to Total Word Ratio
(details)
96.5455%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% words of dialogue between two thieves named Molly and Bill who are planning to break into a children's party to steal all the presents. Only include the prose of the scene without any markdown or commentary.
Model response
Messages
Meta
AI
Molly adjusted the sequined headband itching her left eyebrow, a humiliating unicorn horn for a woman whose resume included the Great Diamond Swap and the Swiss National Bank heist. Beside her, Bill vibrated with the suppressed intensity of a silent phone, clutching a juice box like a detonator, his knuckles white against the orange cardboard. They hid behind a towering papier-mâché castle smelling of glue and despair. Through the translucent orange wall, the target gleamed: a mountain of brightly wrapped parcels on a table of plastic stars. It was the annual bash of six-year-old Timmy the Third, heir to organic crayons and limited-edition plastic dinosaurs. The presents were the real prize. Unopened Lego sets, brand-new tablets, and a rare action figure insured for more than their combined rent. Bill took a shaky sip, leaving a sticky mustache. His eyes darted to the security perimeter, composed of two lollipop-wielding toddlers and a golden retriever in a party hat. The dog looked bored; the toddlers looked dangerous, especially after one projectile vomited on the vicar. Molly checked her glitter bomb watch. Three minutes until the piñata struck, dissolving order into screaming chaos. Molly leaned in and whispered, "Bill, stop drinking the ammo. Save that sugar for the adrenaline crash." Bill lowered the box, his voice trembling. "Do you see that red box, Molly? Under the dinosaur? That is the Holy Grail of toddler commerce. My father died dreaming of that box." "My father died because you tried to picklock a vending machine with a shoelace," Molly hissed. "Focus. Protocol is simple. We drop smoke pellets disguised as confetti cannons. I tackle the dog with a squeaky toy distraction." Bill hissed back, eyes manic. "The laundry basket is checkered. I cannot rob a party in a checkered laundry basket. It says I am an amateur. It says I am a suburban mom who lost her mind." "It says we leave with three thousand dollars in unopened tech gear," Molly countered. "You take the basket, or you risk the kindergartener hug-of-death while hauling Lego bricks." Bill paled. The threat of being hugged to death by children in dinosaur onesies was a psychological failure point flagged in his eval. He nodded, gripping the wicker handle. The basket was a hideous clash of burgundy and beige, screaming discount department store. He stared at it as if it were a curse woven by the dark arts of domesticity, his hand recoiling as the fabric burned his fingertips. "Checkered it is," Bill whispered. "But I am keeping the red box." "Fine. Just do not unwrap the red box. We are thieves, Bill, not savages." Molly pulled the pin on the confetti canister. The party roared. Bill lunged, a whirlwind of polyester and desperation, the checkered basket clattering like war drums. Molly dove for the dog, wielding a rubber chicken like Excalibur. As the confetti cloud exploded, coating Timmy in festive oblivion, Molly smiled. The loot was fresh, and the resistance was sticky.
created_at
"2026-04-27T09:19:28.636Z"
4/27/2026, 9:19:28 AM
model
"qwen/qwen3.6-27b-20260422"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
876
moderation_latency
(null)
generation_time
121827
tokens_prompt
77
tokens_completion
8213
native_tokens_prompt
76
native_tokens_completion
8066
native_tokens_completion_images
(null)
native_tokens_reasoning
7396
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.0290832
router
(null)
provider_responses
0
endpoint_id
"82097366-3a73-4790-ac36-f96ba9f6f577"
id
"chatcmpl-29feeb43-475e-9ee5-86f6-cf1cb2b79847"
is_byok
false
latency
876
model_permaslug
"qwen/qwen3.6-27b-20260422"
provider_name
"Alibaba"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777281568-G2YSPbOO559jBb9t1Rjg"
session_id
(null)
api_type
"completions"
id
"gen-1777281568-TBgPFXNP0N5A1ghvEniG"
upstream_id
"chatcmpl-29feeb43-475e-9ee5-86f6-cf1cb2b79847"
total_cost
0.0290832
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
98.71%
Matches word count
n/a
neededClean
false
words
494
94.38%
Dialogue to Total Word Ratio
Ratio: 31.08%, Deviation: 1.08%
neededClean
false
wordsTotal
502
wordsDialogue
156
96.5455%