NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ8NHNPVJNSVYB7AE24C8N5
Run Details
Model:
Qwen 3 32B
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
7.0s
Cost:
$0.000328
Provider:
AtlasCloud
Evaluation summary
Score
Evaluator
Details Link
90.48%
Matches word count
(details)
0.70%
Dialogue to Total Word Ratio
(details)
45.5935%
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 inflatable chicken costume and hissed, "Bill, if I have to wear this one more minute, I swear I'm going to set it on fire." Bill, zipped inside a neon fanny pack filled with lockpicks and whoopee cushions, grinned. "Relax, it’s only three more hours until the party ends. By then, we'll have every gift wrapped in our getaway van under the bounce house." "That's the plan," Molly said, rolling her eyes. "If the piñatas don't sense us first. Or the bouncy castle doesn’t eat you whole." "First, we pose as entertainers. Then we sabotage the clown so he takes too long. While parents are distracted, we swipe everything under five feet tall and tie string to each box for easy retrieval." "You’re brilliant," Molly sighed. "Also, maybe mentally unstable." Bill shrugged under the fanny pack. “I prefer ‘resourceful.’” A voice crackled over the toy walkie-talkies they'd borrowed from the rental truck. "Mom, my unicorn is melting!" came a kid’s panicked whisper. Bill looked at Molly. "They're buying presents longer than a sandwich now." Molly pulled a mini prybar from her feather skirt. "Let's go make some noise."
created_at
"2026-03-25T10:29:27.866Z"
3/25/2026, 10:29:27 AM
model
"qwen/qwen3-32b-04-28"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1044
moderation_latency
(null)
generation_time
6980
tokens_prompt
93
tokens_completion
305
native_tokens_prompt
76
native_tokens_completion
267
native_tokens_completion_images
(null)
native_tokens_reasoning
0
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
0.000328
router
(null)
provider_responses
0
endpoint_id
"2693bf9a-1732-4008-acc9-d241214e936c"
id
"cb906d5dbefb4927943cb7bf1aeef4a6"
is_byok
false
latency
1044
model_permaslug
"qwen/qwen3-32b-04-28"
provider_name
"AtlasCloud"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1774434567-idv3if90OkxeTKZoeDD7"
upstream_id
"cb906d5dbefb4927943cb7bf1aeef4a6"
total_cost
0.000328
cache_discount
(null)
upstream_inference_cost
0
provider_name
"AtlasCloud"
Evaluation details
Result
Evaluator
Details
Meta Data
90.48%
Matches word count
n/a
neededClean
false
words
190
0.70%
Dialogue to Total Word Ratio
Ratio: 64.92%, Deviation: 14.92%
neededClean
false
wordsTotal
191
wordsDialogue
124
45.5935%