NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ8GY3ATA5NDAZ2AGRTV5GD
Run Details
Model:
Qwen 3 32B
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
11.6s
Cost:
$0.000249
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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 party horn earpiece, sizing up the bounce house. “Bill, if we’re robbing a *children’s birthday*, we need subtlety. Not a taser unicorn.” Bill emerged from a piñata, confetti in his hair. “Subtlety’s overrated. I’ll dress as a balloon artist. No one suspects a guy with a lamb’s head.” “First of all, it’s a *camel*,” Molly said, lobbing a handful of party poppers into her pocket. “Second, you’d need a permit for that lamb. We’re going as clown and… whatever you are when you’re holding a cupcake.” “I’m a *cupcake enthusiast*,” Bill declared, snatching a streamer. “We disable the motion sensors first?” “No sensors. Just a dad with a fog machine and a ‘don’t set off the fire alarm’ sign.” Molly crouched, eyes on the gift table. “We hit when the cake starts. You’ll have 90 seconds between ‘Happy Birthday’ and ‘Why is the smoke alarm going off?’” Bill frowned. “What if we get caught?” Molly grinned. “We’re *characters*. You’ll say it’s part of the theme. ‘Surprise! The magician’s stealing your stuff!’” “魔术!” Bill grinned, then gasped. “Wait, there’s a *pony* here.” “Which is why you don’t pocket the Cabbage Patch dolls. Too many kids yell ‘MY BABY!’” Bill adjusted his hat, now a traffic cone with googly eyes. “What about the *parents*?” Molly handed him a noisemaker. “You blow this. They’ll think it’s the clown being……clouncy.” Bill blew. The thing screeched like a dolphin with a headache. Molly facepalmed. “……or the clown being *dead*. Just. Go.” He sprinted toward the bounce house. “I’ll be the loudest, flattest clown they ever雇用!” Molly sighed, slipping a talking rubber chicken into her bag. “Let’s hope this works. For science. And profit.”
created_at
"2026-03-25T10:26:56.75Z"
3/25/2026, 10:26:56 AM
model
"qwen/qwen3-32b-04-28"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
102
moderation_latency
(null)
generation_time
11495
tokens_prompt
93
tokens_completion
978
native_tokens_prompt
73
native_tokens_completion
868
native_tokens_completion_images
(null)
native_tokens_reasoning
465
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
0.00024888
router
(null)
provider_responses
0
endpoint_id
"6b8c829d-3094-45e7-8139-0a67e09060c3"
id
"cmpl-b26a8cda9239447aab7475bf5ada300f"
is_byok
false
latency
49
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-1774434416-OtqclB9ywAWnQtXOngr7"
upstream_id
"cmpl-b26a8cda9239447aab7475bf5ada300f"
total_cost
0.00024888
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
281
0.00%
Dialogue to Total Word Ratio
Ratio: 65.84%, Deviation: 24.16%
neededClean
false
wordsTotal
281
wordsDialogue
185
0.0000%