NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ853631NDNPPSDZ3GJK220
Run Details
Model:
Qwen3 235B A22B Instruct 2507
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
4.8s
Cost:
$0.000047
Provider:
WandB
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
99.96%
Dialogue to Total Word Ratio
(details)
49.9812%
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
“Are you insane?” Molly whispered, clutching her ski mask like it owed her money. “No,” Bill said, adjusting his belt full of glitter glue sticks. “I’m *inspired*.” “You cannot break into a seven-year-old’s birthday party. That’s not a heist. That’s a war crime.” “It’s not breaking in if the invitation says ‘open bar’—for parents.” “That’s not even the same bar! It’s a juice bar! With tiny napkins!” “Which means security is minimal. Kids cry at loud noises. No alarms. No guard dogs. Just clowns and moral decay.” “Bill, we’ll be mobbed by sugar-crazed gremlins with cupcake breath.” “Then we go in during nap time. Tactical snooze window. Ten minutes, tops.” “And the clown?” “We offer him union rates. He becomes an asset.” “A *clown*? Bill, he has a red nose and a tiny car. That’s not an accomplice, that’s a psychological trigger.” “He drives straight to the getaway van.” “Now you want a *getaway clown*?” “Look, the loot is stacked: bikes, tablets, Build-A-Bears with emotional attachments. That’s emotional leverage. We’re not stealing gifts—we’re liberating them.” “From a piñata?” “Especially from a piñata. Think of the candy! Unmonitored, unregistered, unaccounted-for candy!” “You’re deranged.” “And you’re underdressed. Where’s your clown nose?” “I’m not wearing a nose.” “Then you’re not coming.” “Fine. But if I get attacked by a singing cake, I’m divorcing you.” “Deal. Now grab the face paint. Operation Pin the Theft on the Donkey is a go.”
created_at
"2026-03-25T10:20:28.744Z"
3/25/2026, 10:20:28 AM
model
"qwen/qwen3-235b-a22b-07-25"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
60
moderation_latency
(null)
generation_time
4710
tokens_prompt
77
tokens_completion
446
native_tokens_prompt
72
native_tokens_completion
398
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
0.000047
router
(null)
provider_responses
0
endpoint_id
"c9f927db-d559-49eb-9846-80862c5e3b4e"
id
"chatcmpl-811b9f1cc0540c77"
is_byok
false
latency
25
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-1774434028-wkbvDrH9VLl1JTZcHi76"
upstream_id
"chatcmpl-811b9f1cc0540c77"
total_cost
0.000047
cache_discount
(null)
upstream_inference_cost
0
provider_name
"WandB"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
238
99.96%
Dialogue to Total Word Ratio
Ratio: 91.39%, Deviation: 1.39%
neededClean
false
wordsTotal
244
wordsDialogue
223
49.9812%