NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ8B71N5N992JWR0P90J8AX
Run Details
Model:
Qwen3 235B A22B Instruct 2507
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
3.3s
Cost:
$0.000036
Provider:
WandB
Evaluation summary
Score
Evaluator
Details Link
86.38%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
61.5841%
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
"Bill, are you insane? A kids’ birthday party? With clowns and piñatas and sticky little gremlins?" "Exactly. Chaos, Molly. Pure chaos. No one’s watching the loot when a three-year-old’s having a meltdown over cake." "They’ll have parents. Aunts. Overbearing uncles with grills and judgment." "Then we go as caterers. I’ll wear a hat. You wear a name tag that says ‘Hi, I’m Not Stealing Your Presents.’" "They’ll notice five hundred dollars’ worth of toys just… gone." "Not if we replace them with Monopoly money!" "That’s not even the same emotional value, Bill. Kids want ponies, not fake cash." "Fine, we swap them with expired coupons. Nobody reads the fine print." "Bill, we’re not stealing from a bank. We’re stealing from children. Morally, we’ve hit a new low. Even for us." "Look, I saw the invite—‘Bring your most expensive gift.’ It’s a target-rich environment. And if we’re real heroes, we donate half." "To who? A charity for emotionally scarred toddlers?" "See? You’re already getting into the spirit. Meet me at the party van at five. Wear something festive." "Only if I can bring a juice box." "Only if it’s spiked."
created_at
"2026-03-25T10:23:49.305Z"
3/25/2026, 10:23:49 AM
model
"qwen/qwen3-235b-a22b-07-25"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
61
moderation_latency
(null)
generation_time
2993
tokens_prompt
77
tokens_completion
323
native_tokens_prompt
72
native_tokens_completion
283
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
0.0000355
router
(null)
provider_responses
0
endpoint_id
"c9f927db-d559-49eb-9846-80862c5e3b4e"
id
"chatcmpl-9f23eb81401aa355"
is_byok
false
latency
23
model_permaslug
"qwen/qwen3-235b-a22b-07-25"
provider_name
"WandB"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1774434229-3sN788AxBapLGhiwITrd"
upstream_id
"chatcmpl-9f23eb81401aa355"
total_cost
0.0000355
cache_discount
(null)
upstream_inference_cost
0
provider_name
"WandB"
Evaluation details
Result
Evaluator
Details
Meta Data
86.38%
Matches word count
n/a
neededClean
false
words
189
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
192
wordsDialogue
192
61.5841%