NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ8P53YPTS26RRZAY3YK0VC
Run Details
Model:
Qwen3 235B A22B Instruct 2507
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
13.9s
Cost:
$0.000205
Provider:
Novita
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
84.30%
Dialogue to Total Word Ratio
(details)
42.1500%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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
The glow of a single streetlamp flickered above the minivan parked across from the brightly lit community center. Inside, balloons bobbed to the tune of an off-key rendition of “Happy Birthday,” while Molly adjusted her black ski mask with one hand and scribbled furiously on a notepad with the other. Her pen moved like a caffeinated woodpecker. Bill, crouched beside her, gnawed on a lollipop stolen from the van’s previous owner—a nervous habit he swore only improved his focus. She underlined *cake distraction* three times and circled *clown vulnerability* with particular glee. A diagram emerged: the piñata as both cover and weapon, the bounce house as escape tunnel if inflated westward by 3:17 p.m. precise. Bill pointed mutely at a glitter-covered goat mascot waddling near the snack table, and Molly nodded solemnly, adding “bribe with cupcakes?” with a question mark. They watched a father attempt to operate a bubble machine that instead spat out a single, defeated sputter. Perfect—distractions were multiplying. Molly folded the plan into a paper airplane and nearly launched it toward the roof vent, but Bill gently lowered her arm. Not yet. A chorus of tiny voices shrieked as a toddler opened a box containing a squeaky penguin. Molly’s eyes glistened. Not with sentiment—with logistics. Every gift unwrapped was inventory verified. The van remained silent except for the crinkle of snack wrappers and the occasional, whispered “No face paint. Understood.” The heist would commence post-third round of musical chairs. All according to plan.
created_at
"2026-03-25T10:29:47.78Z"
3/25/2026, 10:29:47 AM
model
"qwen/qwen3-235b-a22b-07-25"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
897
moderation_latency
(null)
generation_time
13732
tokens_prompt
77
tokens_completion
406
native_tokens_prompt
72
native_tokens_completion
343
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
0.00020542
router
(null)
provider_responses
0
endpoint_id
"cf186489-d252-4fb7-aca9-87ef7b557eff"
id
"55881d8a49821fa1783d8599f71d1b91"
is_byok
false
latency
897
model_permaslug
"qwen/qwen3-235b-a22b-07-25"
provider_name
"Novita"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1774434587-n3AdG5YrGi4sN2DGVkKj"
upstream_id
"55881d8a49821fa1783d8599f71d1b91"
total_cost
0.00020542
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Novita"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
249
84.30%
Dialogue to Total Word Ratio
Ratio: 3.57%, Deviation: 6.43%
neededClean
false
wordsTotal
252
wordsDialogue
9
42.1500%