NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-02-25-01KJBFDPSDM73T7NPAMJ0P0JZ5
Run Details
Model:
Qwen 3.5 122B
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
1m 4s
Cost:
$0.035000
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
6.40%
Dialogue to Total Word Ratio
(details)
3.1993%
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 adjusted her black ski mask tightly, though the bright afternoon sun made the disguise laughably ineffective. The suburban driveway stretched before them, lined with pastel balloons that bobbed in the wind like surrender flags. Bill sat in the van, nervously tapping a map against his knee. The stolen minivan smelled of stale pizza and old coffee. The engine idled with a low, anxious hum that vibrated through the floorboards. Molly checked the surveillance photos spread across the dashboard. The house looked peaceful, but she knew the danger lay within the sugar-coated walls. Outside, the neighborhood was quiet, save for the distant sound of a lawnmower cutting through the summer heat. The shadows of the oak trees danced across the hood of the vehicle, casting long, skeletal fingers over the dashboard. Bill shifted uncomfortably, the seatbelt digging into his shoulder. He wiped sweat from his brow, leaving a smudge of grease on his forehead. The air inside the van was thick with tension and the scent of cheap air freshener. “The piñata is a distraction. The real loot is under the tree.” Bill gestured vaguely toward the house. “But what about the cake? It looks like a fortress of sugar.” “We aren’t here for the sugar. We’re here for the consoles.” She finally looked at him. “Are you ready?” “I don’t know. It feels wrong.” “They have insurance. We have rent.” Molly opened the glove box and pulled out two sets of foam finger gloves. “Put these on. We need to blend in.” Bill stared at the foam fingers. “They’re for the Smith family birthday.” “Just wear them.” He sighed, slipping his hands into the oversized plastic digits. “Smile. Nod. Point at the balloons.” “We have a four-minute window before the cake cutting.” “We don’t unwrap. We bag and tag.” “What about the dog?” Molly paused. “There’s a dog?” “I saw a golden retriever in the yard. It’s wearing a party hat.” “That makes it worse.” She grabbed a bag of treats from the dashboard. “Here. Distraction tactic.” Bill took the bag. “It’s a heist, Bill. Ethics are for the people paying us.” He looked at the house again. “What if I get caught?” “Only the invisible kind. The kind made of guilt.” Molly opened the van door. “Stay low. Move fast. Don’t eat the cupcakes.” “Then you tell them you’re a clown.” “Paint one on.” “No promises,” Bill muttered, climbing out. They crept toward the house, foam fingers bobbing. The lawn was a minefield of toys. Molly signaled silence at the patio door. “Ready?” she whispered. “Born ready,” Bill lied. Molly kicked the door open. They were in.
created_at
"2026-02-25T22:41:26.322Z"
2/25/2026, 10:41:26 PM
model
"qwen/qwen3.5-122b-a10b-20260224"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
461
moderation_latency
(null)
generation_time
64155
tokens_prompt
77
tokens_completion
10427
native_tokens_prompt
76
native_tokens_completion
10928
native_tokens_completion_images
(null)
native_tokens_reasoning
10318
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.035
router
(null)
provider_responses
0
endpoint_id
"33d466d1-d2f7-45ed-8ff3-fda7d2e32351"
id
"chatcmpl-ed24a460-359c-9a3d-8796-e8a5e181e003"
is_byok
false
latency
460
model_permaslug
"qwen/qwen3.5-122b-a10b-20260224"
provider_name
"Alibaba"
status
200
api_type
"completions"
id
"gen-1772059286-BGtYckxNVHrMuDPKL6dW"
upstream_id
"chatcmpl-ed24a460-359c-9a3d-8796-e8a5e181e003"
total_cost
0.035
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
439
6.40%
Dialogue to Total Word Ratio
Ratio: 37.41%, Deviation: 7.41%
neededClean
false
wordsTotal
441
wordsDialogue
165
3.1993%