NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHK3B9K3G3XVHPYJS0E2XK6V
Run Details
Model:
Qwen 3.5 397B A17B
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
5m 47s
Cost:
$0.047044
Provider:
Alibaba
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 smoothed the crayon napkin across the dashboard outside. The inflatable castle bounced rhythmically in the wind. Bill adjusted his clown nose, feeling the rubber chafe against skin. He watched a parent chase a toddler, noting the volume of wrapped boxes. "The target window is twenty minutes during the cake service," Molly said. "We move when the singing starts loudly." Bill nodded, wiping grease from his forehead away. He questioned the morality but kept his mouth shut. The smell of stale vanilla hung heavy. Molly opened the back doors to reveal empty burlap sacks. She expected resistance from the magician who looked surprisingly fit. "What if the magician sees us through the mirror trick?" Bill asked. "He knows sleight of hand. It is risky business." Molly dismissed the concern with a wave of her hand. She believed the magician was amateur hour based on reviews. The threat involved the guardian, a golden retriever named Buster. Molly prepared tranquilizer darts disguised as party blowers. She handed one to Bill, who inspected the tube. "Buster eats anything. Even plastic darts," Molly said. "Just blow hard into it." Bill sighed, knowing the dog deserved better than this. He pocketed the dart and checked his oversized suspenders. They dug into his shoulders, reminding him of the humiliation. Molly checked her watch, timing the moment the singing began. It was the cover for noise. "Do we take the cashmere blankets too?" Bill asked. "Those are expensive items." Molly nodded, confirming the blankets were valuable items. She reminded him that luxury goods did not discriminate. The first notes of Happy Birthday drifted out. Molly pulled her clown wig. It itched, but sacrifices were necessary. "Wait, did you bring the bag for the cake?" Bill asked. "I want a slice of chocolate cake." Molly paused, realizing she had forgotten the dessert. She grabbed a trash bag and tossed it to him. It was going to be a messy job. Bill took a breath and stepped toward the castle. "Is the getaway car running?" Bill asked. "We need a fast exit strategy now." Molly turned the key, hearing the engine hum. She signaled for him to move. They were ready to ruin childhoods for profit. "What about the security cameras?" Bill asked loudly. Molly pointed to the blind spot near the slide. She told him not to worry about the feed. It was looped for the hour. "Okay, I am going in now," Bill said nervously. Molly watched him waddle in the shoes. She hoped the plan would work without disaster. The stakes were high for a silly job. "Don't forget the gift cards," Molly called out. "Yes, definitely." Bill waved without turning around his body. He disappeared into the bounce house entrance. Molly waited for the alarm to sound. She checked her phone timers.
created_at
"2026-02-16T11:34:25.344966+00:00"
2/16/2026, 11:34:25 AM
model
"qwen/qwen3.5-397b-a17b-20260216"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
753
moderation_latency
(null)
generation_time
346738
tokens_prompt
77
tokens_completion
12835
native_tokens_prompt
76
native_tokens_completion
13055
native_tokens_completion_images
(null)
native_tokens_reasoning
12443
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
0.0470436
router
(null)
provider_responses
0
id
"chatcmpl-63c0d5f3-8a9f-92c5-b334-599e058f53cf"
status
200
is_byok
false
latency
753
endpoint_id
"180484a2-e24b-495a-ba41-b34b680d5a9f"
provider_name
"Alibaba"
model_permaslug
"qwen/qwen3.5-397b-a17b-20260216"
api_type
"completions"
id
"gen-1771241317-3pAbp0n56ZlRxQxVldJz"
upstream_id
"chatcmpl-63c0d5f3-8a9f-92c5-b334-599e058f53cf"
total_cost
0.0470436
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0007%
Matches word count
n/a
neededClean
false
words
467
5.0398%
Dialogue to Total Word Ratio
Ratio: 22.27%, Deviation: 7.73%
neededClean
false
wordsTotal
467
wordsDialogue
104
2.5203%