NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1H9P2VFN200F4140A6VTD54
Run Details
Model:
Qwen 3.8 Max (Reasoning, XHigh)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
4m 18s
Cost:
$0.081020
Provider:
Alibaba
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 spread the birthday invitations across the van floor like tiny colorful maps leading to a very sugary crime scene. Bill wore a red clown nose he had found in the glove box and refused to explain it to anyone. The party house glowed three blocks away, full of balloons, cake, and presents stacked beneath a bright paper rainbow outside. Molly tapped her pencil against a drawing of the yard while Bill polished a spoon with suspiciously focused dedication now. Their van smelled like old rope, cheap candles, and a banana that had once belonged to a very small clown. "We will enter through the bouncing castle and grab every single gift from the table," Molly said while smiling widely. Bill nodded as if this were a serious military operation and not a plan involving children and sticky frosting everywhere. "What about the parents? They usually carry cameras, juice boxes, and terrifying opinions about strangers," Bill whispered, trembling slightly again. Molly drew a line around the cake table and labeled it with a dramatic arrow and three gold stars nearby. "Parents can be distracted with free samples of crackers and a puppet show about sharing," Molly said, drawing tiny arrows. Bill raised one finger, then lowered it because he could not remember whether questions were allowed during very serious heists. "If a child cries, we freeze like garden gnomes until a magician appears to help," Molly said, practicing her pose. Bill imagined himself disguised as a magician, then remembered he had only one rabbit and it was very cheap rubber. "I will wear the chicken costume, cluck and carry gifts to the van in daylight," Bill said, saluting badly twice. Molly considered this, then decided that arguing with Bill about poultry was a waste of very precious criminal planning time. "We need a signal, perhaps three honks, one sneeze, and somebody yelling about missing cupcakes," Molly said, honking softly once. Bill practiced the sneeze, which sounded like a small trombone falling down stairs inside a dusty blanket fort that afternoon. "What if the pinata is guarding the presents with unexpected candy reinforcements for us all?" Bill asked, eyes wide now. Molly erased three incorrect stars and replaced them with a drawing of a stick figure running very fast away now. "If the pinata attacks, I will compliment its ribbon and slowly back toward the exit," Bill said, backing away dramatically. Outside, a bicycle bell rang, and both thieves ducked beneath a pile of gift bags printed with little yellow ducks. "This is exactly the sort of danger that requires stealth, silence, and a large pocket," Molly whispered, adjusting her gloves. Bill checked his pockets, which contained string, a whistle, two crackers, and an instruction manual for something possibly useful later. "Remember, we are not monsters. We leave one present for the smallest child, then flee," Bill said, blinking nobly twice. They bumped fists, put on party hats, and rolled toward the backyard like two very determined dessert enthusiasts on duty.
created_at
"2026-09-02T14:52:03.319Z"
9/2/2026, 2:52:03 PM
model
"qwen/qwen3.8-max-20260803"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
869
moderation_latency
(null)
generation_time
257902
tokens_prompt
75
tokens_completion
11858
native_tokens_prompt
127
native_tokens_completion
13461
native_tokens_completion_images
(null)
native_tokens_reasoning
12853
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
0.08102
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788360723-KfSrUb6YkdX3RzDqLfUO"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788360723-oP8XnGtvx3sS7VKrtgdQ"
upstream_id
"chatcmpl-4a632587-f524-9ec4-9b44-53fd20ab1a45"
provider_responses
0
endpoint_id
"6332e9fd-e0c4-4cae-8c17-d25bc774864a"
id
"chatcmpl-4a632587-f524-9ec4-9b44-53fd20ab1a45"
is_byok
false
latency
869
model_permaslug
"qwen/qwen3.8-max-20260803"
provider_name
"Alibaba"
status
200
total_cost
0.08102
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
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%