NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ7QRFQJBK2QGX3P696EMEE
Run Details
Model:
Qwen3 235B A22B Instruct 2507
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
4.2s
Cost:
$0.000040
Provider:
WandB
Evaluation summary
Score
Evaluator
Details Link
1.04%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
18.9120%
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
“Molly, this is the dumbest plan you’ve ever had. And last week you tried to rob a mime convention.” “It’s not dumb! It’s genius! Who expects thieves at a five-year-old’s birthday party?” “Literally anyone with common sense. Also, clowns. Clowns are watching.” “They’re not real clowns, Bill. They’re underpaid college students in face paint.” “And tiny hats. Don’t forget the tiny hats. It’s humiliating.” “We wear the hats, Bill. That’s the disguise.” “You want me to wear a glittery red nose and steal presents from toddlers?” “Not steal. *Relocate*. We’re redistributing joy.” “Joy? There’s a piñata shaped like a unicorn with asthma. That’s not joy, that’s trauma.” “Look, the parents are distracted, the cake’s got enough sugar to power a small country, and the security is a guy named Dave who fell asleep by the juice fountain.” “And the kids?” “They’ll cry for like… twelve minutes. Then someone’ll hand them a sticker and they’ll move on. Kids are resilient.” “What if one of them recognizes me?” “From what? The nightly news? Bill, you once returned a stolen lamp because it didn’t match your couch.” “That was a principled stand!” “Just grab the presents. Stick to the plan. In, out, no eye contact with anyone under four feet tall.” “…Do I at least get to keep the goody bag?” “No.” “But it’s got *crayons*.” “Focus, Bill. Focus.”
created_at
"2026-03-25T10:13:12.068Z"
3/25/2026, 10:13:12 AM
model
"qwen/qwen3-235b-a22b-07-25"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
139
moderation_latency
(null)
generation_time
3761
tokens_prompt
77
tokens_completion
417
native_tokens_prompt
72
native_tokens_completion
332
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
0.0000404
router
(null)
provider_responses
0
endpoint_id
"c9f927db-d559-49eb-9846-80862c5e3b4e"
id
"chatcmpl-b3f9fad2bd0f09dc"
is_byok
false
latency
31
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-1774433592-ersH7m7mWkSUd5MUm0gN"
upstream_id
"chatcmpl-b3f9fad2bd0f09dc"
total_cost
0.0000404
cache_discount
(null)
upstream_inference_cost
0
provider_name
"WandB"
Evaluation details
Result
Evaluator
Details
Meta Data
1.04%
Matches word count
n/a
neededClean
false
words
226
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
228
wordsDialogue
228
18.9120%