NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-27-01KQ74F9PMK2M7Q2VS1JWT0HCK
Run Details
Model:
Qwen 3.6 27B
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
2m 57s
Cost:
$0.039617
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
96.84%
Dialogue to Total Word Ratio
(details)
48.4218%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% 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
The van smelled of stale pretzels and poor decisions. Molly adjusted her binoculars, tracking the suburban mansion while Bill wrestled a grappling hook made of spoons. Dust motes danced in the sunlight cutting across the dashboard, illuminating the emergency ration of sour patch kids. Molly shifted in the cracking leather seat, recalibrating the focus on the target. "Target locked," Molly muttered. "Birthday Kid. Age seven. Sugar crash imminent. We hit the presents when the piñata breaks, and we don't leave a single wrapper behind." Bill squinted through the window, his elbow knocking a rattle through the cabin that sounded like dying seagulls. A golden retriever trotted closer, sniffing the tires with aggressive curiosity. Bill froze, watching the dog's snout twitch. "Shark piñata has teeth like a tax auditor. Absolutely terrifying teeth. And drones? Why are there so many damn drones?" "Aunt Linda is live-streaming," Molly said, tapping a stained map. "One breach and we're cancelled by the PTA. We need a loop." "I can loop the feed, but I'll need the router," Bill replied, grease smearing his knuckles. "Router is inside. Near the cake." "Cake is the anchor," Molly agreed, dropping the binoculars. "Eyes go to sugar, hands go to loot. I've packed the smoke canisters with baby powder. Parents will sniff; we'll swipe." Molly reached into her coat, producing a remote wrapped in duct tape. Her finger hovered over the red button. The heist's fate rested on a wireless signal punching through drywall and a microwave. "Classic," Bill nodded, patting his vest. "Legos for me. High liquidity. Easy to fence." "Legos?" Molly glared, checking her satchel. "I claim tablets and cash cards. Legos are just sharp bricks. I need screens." Bill's stomach growled, echoing in the cramped cabin. He suppressed a wince, adjusting his cap. The sweat made his palms slippery on the steering wheel. "Fine. You take the tech," Bill conceded. "I'll take the robots. Vintage robots buy silence from cops." "Deal," Molly checked her watch. "Leave the giant bear. Too bulky. We're thieves, not monsters. Though your spoon-hook argues otherwise, and frankly, I'm worried about its structural integrity." Bill defended his weapon, holding the spoons aloft. "Spoon-hook is elegant! Silent, musical, and it serves salad at the getaway." "It serves as a reason I might arrest you myself," Molly snapped, sliding the map away. "Hook ready. We move on my mark." "Hook ready," Bill whispered, gripping the metal until his knuckles whitened. "Promise me. If caught, I blame you for the fondue fork, absolutely the fondue fork." Molly threw the van into reverse. "Deal. But remember, I wore the ugly cardigan."
created_at
"2026-04-27T09:31:35.514Z"
4/27/2026, 9:31:35 AM
model
"qwen/qwen3.6-27b-20260422"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1260
moderation_latency
(null)
generation_time
177116
tokens_prompt
77
tokens_completion
9519
native_tokens_prompt
76
native_tokens_completion
10992
native_tokens_completion_images
(null)
native_tokens_reasoning
10349
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
0.0396168
router
(null)
provider_responses
0
endpoint_id
"82097366-3a73-4790-ac36-f96ba9f6f577"
id
"chatcmpl-d27e5116-4f58-99cc-827b-251d317409c8"
is_byok
false
latency
1260
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-1777282295-FTYg0C42FDKv4Q2ytzK4"
session_id
(null)
api_type
"completions"
id
"gen-1777282295-5eS50gOSgzej41t2migJ"
upstream_id
"chatcmpl-d27e5116-4f58-99cc-827b-251d317409c8"
total_cost
0.0396168
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
434
96.84%
Dialogue to Total Word Ratio
Ratio: 50.80%, Deviation: 0.80%
neededClean
false
wordsTotal
437
wordsDialogue
222
48.4218%