NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ7VB0MWEYZVC3DWTTG15MN
Run Details
Model:
Qwen3 235B A22B Instruct 2507
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
9.1s
Cost:
$0.000184
Provider:
Novita
Evaluation summary
Score
Evaluator
Details Link
0.49%
Matches word count
(details)
92.64%
Dialogue to Total Word Ratio
(details)
46.5651%
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
Rain drummed steadily on the corrugated tin awning of the abandoned taco truck they were using as a hideout. Inside, crumpled maps, a toy binoculars with one lens missing, and a half-eaten fruit gusher littered the counter. Molly, wiry and intense, had spread a birthday party seating chart across the damp linoleum, held down by a ketchup bottle and a squirrel figurine stolen from a dentist’s waiting room. She squinted at the diagram, tapping a red dot labeled "Cake Table – HIGH RISK" with the eraser end of a pencil. Bill sat slumped on an overturned bucket, attempting to balance a rubber chicken on his head while reviewing security patterns via a child’s spy watch that only played circus music when activated. The party was for seven-year-old Trevor Wilbanks, heir to the Wilbanks Yogurt Fortune (mini spoons only), and the guest list included no less than three off-duty cops pretending to be clowns. Molly had already tested the perimeter disguised as a balloon artist named "Twisty" and came back with two contorted poodles and a mild hatred of squeaky noises. They had two minutes between piñata frenzy and the ice cream truck’s arrival—precisely when parental attention would peak at sugar-related disaster prevention. Bill adjusted the rubber chicken. Molly exhaled sharply. This was either their masterpiece or a federal misdemeanor involving glitter. “Are we really doing this?” “Yes.”
created_at
"2026-03-25T10:15:09.08Z"
3/25/2026, 10:15:09 AM
model
"qwen/qwen3-235b-a22b-07-25"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1396
moderation_latency
(null)
generation_time
9085
tokens_prompt
77
tokens_completion
369
native_tokens_prompt
72
native_tokens_completion
306
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
0.00018396
router
(null)
provider_responses
0
endpoint_id
"cf186489-d252-4fb7-aca9-87ef7b557eff"
id
"ae72064318cf856f70cf3e5973ca2ca7"
is_byok
false
latency
1396
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-1774433709-cgMdHLqHgTRsbA15Zqz4"
upstream_id
"ae72064318cf856f70cf3e5973ca2ca7"
total_cost
0.00018396
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Novita"
Evaluation details
Result
Evaluator
Details
Meta Data
0.49%
Matches word count
n/a
neededClean
false
words
227
92.64%
Dialogue to Total Word Ratio
Ratio: 4.74%, Deviation: 5.26%
neededClean
false
wordsTotal
232
wordsDialogue
11
46.5651%