NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-22-01KS86EJ98NNAFA64N383R59VQ
Run Details
Model:
Qwen3.7 Max
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
7m 4s
Cost:
$0.119320
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 tapped her pen against the blueprint, scanning the heavily guarded suburban perimeter. The dining room was marked with a red circle, indicating the primary loot zone. Bill leaned over her shoulder, squinting at the backyard layout. He pointed at a massive yellow structure near the patio doors, his heavy brow furrowed deeply. "The primary objective is the dining room table." Molly sighed, tracing a route along the side of the house. She knew the inflatable obstacle would be a major problem. Bill crossed his arms, clearly unhappy about the entertainment staff stationed inside the residence. He hated dealing with unpredictable elements wearing oversized shoes. "What about the bouncy castle blocking the rear exit?" "We bypass the inflatable structure and enter through the kitchen." Molly offered a reassuring smile. She reached into her duffel bag and pulled out a brightly colored packet of long rubber tubes. Bill stared at the supplies in absolute horror, shaking his head. He was a professional safecracker, utterly refusing to compromise his professional dignity for this job. "There will be a clown guarding the hallway." "Clowns are easily distracted by balloon animals." "I am not making a poodle out of latex." "You will make whatever shape neutralizes the threat." Molly rolled her eyes, tapping the blueprint to emphasize the urgency. She explained that stealth was their only real advantage. Bill shifted his weight, glancing nervously toward the window as if expecting angry toddlers. He worried about the many witnesses, knowing full well that young children were brutally honest. "What if the kids see us taking the gifts?" "We wear the mascot costumes stored in the garage." "Which mascots?" "A giant purple dinosaur and a smiling hotdog." Molly grinned widely, revealing her brilliant contingency plan. She pointed toward the detached garage on the map. Bill looked relieved but immediately questioned the specific disguises available. Molly described the outfits in vivid detail, watching his expression drop into a state of sheer absolute misery. "I refuse to be the hotdog." "The dinosaur suit lacks peripheral vision." "Fine but I keep the mustard packet." Bill protested loudly, refusing to wear the humiliating food costume. Molly countered his argument with practical logic regarding visibility limitations. Bill grumbled in defeat, accepting his rather unfortunate culinary fate today. "Listen Bill. The gifts are wrapped in fragile paper." "I brought my tactical scissors." "Grab the boxes and run to the van." Molly snapped her fingers, bringing his attention back to the delicate stolen goods. Bill nodded confidently, patting the specialized cutting tool in his belt. Molly outlined the primary escape route. Bill hesitated, suddenly troubled by the inevitable emotional fallout of their crime. "What if someone cries?" "We leave a decoy box of Brussels sprouts." "That is wonderfully evil." "I learned it from my grandmother." Molly smirked confidently, proudly producing a small, neatly wrapped substitute decoy package. Bill chuckled loudly, deeply impressed by her absolutely ruthless dedication to psychological warfare. Molly winked playfully, proudly admitting the true origin of her highly devious strategy.
created_at
"2026-05-22T15:59:49.824Z"
5/22/2026, 3:59:49 PM
model
"qwen/qwen3.7-max-20260520"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
873
moderation_latency
(null)
generation_time
241139
tokens_prompt
77
tokens_completion
12435
native_tokens_prompt
76
native_tokens_completion
15884
native_tokens_completion_images
(null)
native_tokens_reasoning
15237
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
0.11932
router
(null)
provider_responses
0
endpoint_id
"6d30fdc6-f2a7-4bdf-b6ba-32353496d723"
id
"chatcmpl-800af9ba-519f-9d78-ac52-eefa8e8c1b42"
is_byok
false
latency
873
model_permaslug
"qwen/qwen3.7-max-20260520"
provider_name
"Alibaba"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.15.0; linux; x64))"
http_referer
(null)
request_id
"req-1779465589-uD9qpq8KKxlZ9TrhYszD"
session_id
(null)
api_type
"completions"
id
"gen-1779465589-FDNswmYUXOdfBxKKWyP8"
upstream_id
"chatcmpl-800af9ba-519f-9d78-ac52-eefa8e8c1b42"
total_cost
0.11932
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
response_cache_source_id
(null)
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%