NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ7ZGHXYS0P4MXXV8EGVYFB
Run Details
Model:
Qwen 3 32B
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
10.1s
Cost:
$0.000221
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
99.06%
Dialogue to Total Word Ratio
(details)
49.5323%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 50% 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 trench coat inside the unmarked van, eyeing the suburban home through the rear window. Streamers looped like neon spaghetti across the living room, and the lawn thumped with a bounce house shaped like a cartoon dinosaur. “This is a nightmare,” Bill muttered, fiddling with a lock-picking kit. “We’re supposed to steal presents, not get pepper-sprayed by a toddler with a water gun.” “Relax,” Molly said, peering at the guest list on her phone. “The parents are all inside drinking punch laced with rum. We’ll slip in, disable the motion sensors, and vanish before the piñata’s empty.” Bill gaped. “Motion sensors? You said *children’s party*, not *Mission: Impossible*.” “Eh, I added the sensors.” Molly grinned. “Makes it fun.” He groaned. “And the ‘getaway plan’ is…?” “Squirrel costume.” She nodded toward a garish green outfit in the back. “We’ll blend in with the party favor chaos.” “Chaos? Molly, this is a *kids’* party! They’ll think you’re a walking dessert table and try to feed you.” A car pulled into the driveway. Molly shushed him as the door swung open, revealing a babysitter in a tiara clutching a tub of candy. “Phase one: sabotage the clown’s script,” Molly whispered. “We need ten minutes of *controlled* panic.” Bill saluted, then winced. “What if we trigger a glitter bomb?” “Then we’ll leave a trail of sparkle and run like Olympians.” “Brilliant. Nothing says ‘stealth’ like iridescent footprints.” Molly kicked the van’s door. “Move your Daisy Duke posterior. The cake isn’t going to frost itself.” Bill mumbled something about cake puns as they slunk toward the house, the bounce house emitting a mortified squeal.
created_at
"2026-03-25T10:17:25.826Z"
3/25/2026, 10:17:25 AM
model
"qwen/qwen3-32b-04-28"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
137
moderation_latency
(null)
generation_time
9945
tokens_prompt
93
tokens_completion
897
native_tokens_prompt
73
native_tokens_completion
769
native_tokens_completion_images
(null)
native_tokens_reasoning
411
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
0.00022116
router
(null)
provider_responses
0
endpoint_id
"6b8c829d-3094-45e7-8139-0a67e09060c3"
id
"cmpl-4d2c53807b8b4de5aa8d6978072410f3"
is_byok
false
latency
67
model_permaslug
"qwen/qwen3-32b-04-28"
provider_name
"DeepInfra"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1774433845-9a4udJhOZDEa5VFlkORT"
upstream_id
"cmpl-4d2c53807b8b4de5aa8d6978072410f3"
total_cost
0.00022116
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
271
99.06%
Dialogue to Total Word Ratio
Ratio: 53.11%, Deviation: 3.11%
neededClean
false
wordsTotal
273
wordsDialogue
145
49.5323%