NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-22-01KS862HQPD4HXP318J7QQNBT2
Run Details
Model:
Qwen3.7 Max
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
2m 16s
Cost:
$0.071027
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
43.38%
Matches word count
(details)
8.52%
Dialogue to Total Word Ratio
(details)
25.9479%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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 the crayon drawing with her pen, tracing the property lines carefully. "We go in through the bounce house." Bill crossed his arms, glaring at the childish artwork. "The bouncy castle is a tactical nightmare. Static electricity will ruin my comms." "You are not wearing comms to a seven year old birthday party, Bill. You are wearing a clown nose." He rubbed his forehead, clearly regretting his life choices. "I refuse to wear the nose. It compromises my peripheral vision." "It is a red foam nose. You will wear it and smile. The presents are stacked near the punch bowl." Bill leaned over the table, studying the crude layout. "What kind of security are we looking at? Are there dogs?" "It is a suburban backyard. The only dog is a golden retriever named Barnaby who is afraid of thunder." He patted his jacket pocket confidently. "I brought liver treats just in case. What about the parents?" Molly sipped her cold coffee, ignoring his dramatic sighs. "Distracted. They will be drinking mimosas and complaining about property taxes. That gives us a twelve minute window during the magic show." "A magic show? Please tell me we are not stealing from a magician. Those guys are paranoid." She rolled her eyes, used to his endless complaining. "It is a guy named Marvelous Marvin. His best trick is pulling a damp dove out of a sleeve. He is not a threat." Bill sighed. "Fine. I will secure the perimeter near the petting zoo while you bag the loot." "You cannot go near the petting zoo. Last time you got distracted by a goat." He shifted uncomfortably, remembering the aggressive farm animal. "That goat had my lockpicks. I was retrieving company property." "Just stick to the plan. We grab the wrapped boxes, toss them over the fence into the van, and extract before the piñata breaks." Bill raised an eyebrow. "What if the piñata breaks early?" "Then we run. Sugar crazed toddlers are faster than they look." He leaned back. "I still think we should just rob a bank." "Banks have vaults. This party has a cardboard cutout of Spiderman. We are taking the easy money." Bill picked up a wrapped gift from the corner of the room. "These presents are mostly educational toys, Molly. The resale value on a beginner chemistry set is terrible." "Timmy is getting a drone. A very expensive drone. And I already have a buyer on the dark web." He laughed out loud. "A buyer for a stolen toy drone?" "Do not judge my contacts. Just be ready to drive." Bill grabbed his car keys from the table. "I will have the engine running. But if a kid asks me to make a balloon animal, I am aborting the mission." "You do not know how to make balloon animals anyway." He smiled and opened the door. "Exactly. It would blow my cover."
created_at
"2026-05-22T15:50:12.996Z"
5/22/2026, 3:50:12 PM
model
"qwen/qwen3.7-max-20260520"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1049
moderation_latency
(null)
generation_time
135988
tokens_prompt
77
tokens_completion
7438
native_tokens_prompt
76
native_tokens_completion
9445
native_tokens_completion_images
(null)
native_tokens_reasoning
8802
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
0.0710275
router
(null)
provider_responses
0
endpoint_id
"6d30fdc6-f2a7-4bdf-b6ba-32353496d723"
id
"chatcmpl-aa83307f-f7c3-9eac-a0b9-9339ff3f5afd"
is_byok
false
latency
1049
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-1779465013-4yrGUHHkIWKtZ7vMYSHD"
session_id
(null)
api_type
"completions"
id
"gen-1779465013-Qur9iRLWdYbfw1S63Iy0"
upstream_id
"chatcmpl-aa83307f-f7c3-9eac-a0b9-9339ff3f5afd"
total_cost
0.0710275
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
43.38%
Matches word count
n/a
neededClean
false
words
483
8.52%
Dialogue to Total Word Ratio
Ratio: 77.02%, Deviation: 7.02%
neededClean
false
wordsTotal
483
wordsDialogue
372
25.9479%