NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ8F4HMV6H6R2NWG30YQPZV
Run Details
Model:
Qwen3 235B A22B Instruct 2507
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
4.5s
Cost:
$0.000046
Provider:
WandB
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 50% 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 adjusted her black ski mask, which kept slipping over one eye. "You really think a kids’ birthday party is our best target?" Bill, crouched behind a shrub, tugged at his too-tight jumpsuit. "Free rein, zero security, and a pile of unattended gifts! It’s like Christmas threw up in a bounce house." "Most of those gifts are toy dinosaurs and glitter glue, Bill." "So? Resell value’s low, but have you seen the resale market for unused crayons? Surprisingly niche." Molly sighed. "We’re not stealing glitter glue." "Who’s stealing? We’re *relocating* with intent to… appreciate." "Appreciate?" "Okay, maybe resell to a third-grade black market. Look, the clown’s on break, the parents are on wine time, and the piñata’s still full. We go in, swipe the loot, vanish like a dad at snack time." "And what about the kids?" "Kids love a good mystery! We’ll be urban legends. The Gift Goblins. Sounds better than ‘Molly and Bill, Cat Burglars.’" "That’s not even our name!" "Could be! Also, I brought the sack." Molly stared at the red velvet Santa bag he produced. "That’s from your apartment Christmas." "It’s festive! Adds authenticity." "You’re ridiculous." "And you love it. Now, ready? On three. One—" "Wait! I just saw a tiny child eating a cupcake with your face drawn on it. That’s ominous." "Even more reason to leave quickly. Two—" "Bill." "Three!" They lunged forward—only to trip over a rogue scooter and crash into a stack of juice boxes. The sprinklers activated. The music skipped. A toddler pointed and screamed, “Mommy! Santa’s fighting!”
created_at
"2026-03-25T10:25:57.816Z"
3/25/2026, 10:25:57 AM
model
"qwen/qwen3-235b-a22b-07-25"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
71
moderation_latency
(null)
generation_time
4430
tokens_prompt
77
tokens_completion
442
native_tokens_prompt
72
native_tokens_completion
390
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
0.0000462
router
(null)
provider_responses
0
endpoint_id
"c9f927db-d559-49eb-9846-80862c5e3b4e"
id
"chatcmpl-bed6ff63229e40ed"
is_byok
false
latency
29
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-1774434357-yFcZsxhN0VTnOM0HID8Y"
upstream_id
"chatcmpl-bed6ff63229e40ed"
total_cost
0.0000462
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
258
0.00%
Dialogue to Total Word Ratio
Ratio: 75.38%, Deviation: 25.38%
neededClean
false
wordsTotal
260
wordsDialogue
196
0.0000%