NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-27-01KQ738B67D5XT2AY5Z4M81ZD0
Run Details
Model:
Qwen 3.6 27B
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
1m 47s
Cost:
$0.021428
Provider:
Morph
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
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 damp alley smelled of wet cardboard and discarded glitter. Molly adjusted her tactical visor, peering through the slats of a plastic playhouse. Bill was currently wrestling with a folding chair, trying to make it look like a convincing tree. His breathing echoed off the brickwork. We need to rethink the approach, Bill groaned, abandoning the chair and wiping confetti from his eyebrows. The perimeter is too tight. I do not handle well in open spaces. It is a suburban driveway, Bill, Molly replied, tapping her clipboard with a yellow highlighter. The only security we have to worry about are a few well-meaning mothers and a golden retriever that chases its own tail. We go through the garden gate, grab the boxes, and bounce before the cake arrives. And what if the host mother has one of those motion-activated sprinklers? Bill asked, finally untangling his sleeves from a bunch of plastic party hats. I am not built for hydration. The moss will mildew. The entire operation collapses into a damp, floral singularity. You are wearing a craft store discount section, not actual foliage, Molly pointed out, checking her wristwatch for the time. And the presents are stacked in a pyramid behind the folding table. We hit the gap, grab the cardboard, and load them into the stroller. The stroller is for one, not forty-three, Bill countered, pointing at the sketch. Do you have any idea how much bulk is in boxed board games and plastic farm animals? We will need a cargo van, a forklift, and possibly a structural engineer. We will need three trips and a very convincing toddler meltdown, Molly said, tapping the map where she had marked the prize locations with red sharpie. Grab the big boxes first, leave the socks and the bath toys as a distraction. I object to prioritizing the socks, Bill said, crossing his arms, which caused a plastic sunflower to dangle dangerously close to his eye. The bath toys are solid. The rubber duck has an alarm system. I tested it with a paperclip. You set off the neighbor's car alarm and got fined sixty dollars, Molly reminded him. Focus. We go in, we grab the presents, we get out before the cake arrives. The cake is our time bomb. Sugar makes children unpredictable. I prefer the precision of the stroller method, Bill insisted, smoothing his collar. We roll in, we take the mountain, we roll out. It is elegant. It has wheels. Wheels do not scream. Wheels do not have cup holders, Molly countered. How do you plan to carry the loot in a cup holder? I did not plan to carry them in a cup holder, Bill said, pointing at the map. I planned to use them as a distraction while a larger vehicle arrives. There will not be a larger vehicle, Molly said. The van is being towed for unpaid parking tickets. Then we go with the stroller, Bill declared. But we take the bath toys first. Bill, they are rubber frogs and a plastic castle that requires three AA batteries and a parental warning label. Exactly, Bill nodded. High value, low risk. I can work with that. Molly sighed, pinching the bridge of her nose. We are stealing presents from a suburban block party, not the crown jewels. Just grab everything in the pile. I am trying to maintain professional standards, Bill muttered, adjusting his ghillie daisies. What about the wrapping paper? Leave it, Molly said. We do not need two hundred pounds of flimsy cardboard. Are you sure? Bill asked. In my experience, cardboard is the foundation of every successful extraction. The foundation is us not getting tackled by a ten-year-old with a juice box and a death grip. Then I demand a tactical juice box, Bill said, pointing at the driveway. I need a beverage that contains enough sugar to fuel a quick exit strategy. Noted, Molly said. We get the boxes, we get out, and we do not touch the juice table. And the pinata? Bill pressed, tugging at a stubborn plastic petal. Do we dismantle it first or leave it as cover? Leave it alone, Molly said. It is full of cheap candy and a single plastic ring pop. It is not a cover, it is a hazard. Understood, Bill said, finally free from the floral contraption. I will prepare the stroller for heavy lifting. Bill, Molly said, watching a parade of rainboots trample across the lawn. Do not overengineer a bag of board games. Never, Bill replied, stepping out from behind the dumpster and vanishing into the shadows of a parked minivan. The evening air carried the scent of artificial cherry and wet pavement. Somewhere nearby, a bicycle bell rang twice. Molly closed her folder and adjusted her grip on the clipboard. She knew this would go sideways. It always did. Still, the plan solid enough for Tuesday.
created_at
"2026-04-27T09:10:19.084Z"
4/27/2026, 9:10:19 AM
model
"qwen/qwen3.6-27b-20260422"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
960
moderation_latency
(null)
generation_time
106557
tokens_prompt
77
tokens_completion
8934
native_tokens_prompt
76
native_tokens_completion
8911
native_tokens_completion_images
(null)
native_tokens_reasoning
7754
native_tokens_cached
64
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
0.0214282
router
(null)
provider_responses
0
endpoint_id
"826cab4e-dc26-4472-83be-e4ddd5ebd2ab"
id
"c01da2311eb24c6d899c108e710ac55f"
is_byok
false
latency
766
model_permaslug
"qwen/qwen3.6-27b-20260422"
provider_name
"Morph"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777281019-Frqq8m26B07shFVZOOsD"
session_id
(null)
api_type
"completions"
id
"gen-1777281019-MMzIdm6uB0xHeTb9XOtu"
upstream_id
"c01da2311eb24c6d899c108e710ac55f"
total_cost
0.0214282
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Morph"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
815
0.00%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 50.00%
neededClean
false
wordsTotal
820
wordsDialogue
0
0.0000%