NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KXK8HW14H1RS5YKJV6T9VB
Run Details
Model:
Qwen 3.8 27B (Reasoning, XHigh)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
1m 39s
Cost:
$0.022524
Provider:
Parasail
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
100.0000%
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 crouched behind a parked tricycle, her cap pulled low over one ear. She tapped a folded map onto Bill's knee, where balloons had been sketched in red. "Bill, the party is on Saturday, and every present is wrapped in glitter like a promise." Bill adjusted his oversized tie, a thing that had once belonged to a wedding singer. "If we move like dust bunnies, nobody will hear us." The map showed the backyard, a slide, a table full of boxes, and a mother guarding lemonade with the focus of a border guard. "The host will be serving cake, so we must be in and out before the piñata." Bill unfolded a napkin and drew a cart, then a window, then a very small hat. "My plan is simple. We wear clown shoes, look ridiculous, and take the gifts." He pointed at the window with a grape. "Then I will open the window while you distract the parent with a tiny hat." Molly raised one finger, as if the finger itself might file a complaint. "A tiny hat is not a full strategy." Bill let the grape roll across his palm. "It is if it comes off on the principal." Molly did not laugh, which was suspicious, because Bill had once described a hat as tactical furniture. She remembered a previous job, a museum after midnight, when a penguin exhibit had delayed them by forty seconds and a velvet rope had become the most emotional object in the building. Bill had apologized to the rope, and Molly had not corrected him. Now the backyard map spread between them, and the piñata, a lumpy dragon wearing cardboard teeth, seemed to watch them through a hole in its side. The air was thick with the promise of juice boxes and the first balloons had already begun to arrive, like polite, colorful spies while the street lamp blinked its approval. "Ridiculous is the point, Bill. The children will think we are part of the show." They sat on the curb, two professional criminals in shirts that had survived several laundromat mysteries. The street smelled of cut grass and someone's questionable barbecue. Molly counted the necessary tools in her mind, then said the count out loud. "Okay. We need gloves, tape, a cart, and a very convincing cough." A yellow dog trotted past, tail high, asking nothing and promising everything. "What about the dog? Dogs smell bad intentions and birthday cake." Bill nodded, because Bill believed in solutions he had not yet explained. "We give him a bone made of rope and call it a souvenir." A marching band began somewhere behind the laundromat, and the plan tightened around the music. Molly checked her shoes, which were already too loud. Bill checked his hat, which was already too wide. "And if the fire alarm screams, we blame the sprinkler, loudly." They stood together as the sun moved across the fence, and the party, still a day away, looked like a mountain made of ribbons.
created_at
"2026-09-03T15:18:31.401Z"
9/3/2026, 3:18:31 PM
model
"qwen/qwen3.8-27b-20260814"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
248
moderation_latency
(null)
generation_time
98996
tokens_prompt
75
tokens_completion
8451
native_tokens_prompt
118
native_tokens_completion
10225
native_tokens_completion_images
(null)
native_tokens_reasoning
9595
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"
service_tier
(null)
usage
0.0225245
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788448711-0o7dwAZSvyz7NfPbKEaH"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788448711-pc9jElWSBUtlYk8Ex15N"
upstream_id
"chatcmpl-0cacaa8ed6501b9836ce222f0df677f4"
provider_responses
0
endpoint_id
"a58a4a1f-4e83-4776-a40e-9f55b72783cb"
id
"chatcmpl-0cacaa8ed6501b9836ce222f0df677f4"
is_byok
false
latency
244
model_permaslug
"qwen/qwen3.8-27b-20260814"
provider_name
"Parasail"
status
200
total_cost
0.0225245
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Parasail"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
500
100.00%
Dialogue to Total Word Ratio
Ratio: 30.00%, Deviation: 0.00%
neededClean
false
wordsTotal
500
wordsDialogue
150
100.0000%