NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ7SQMPSVRNSYQQ14Z1PF9C
Run Details
Model:
Qwen3 235B A22B Instruct 2507
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
5.5s
Cost:
$0.000047
Provider:
WandB
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
99.73%
Dialogue to Total Word Ratio
(details)
49.8652%
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 adjusted her ski mask. “You’re telling me the only time the clown’s not guarding the loot is during the ‘Pin the Tail on the Donkey’ game?” Bill nodded, checking his watch. “It’s a two-minute window, tops. The clown has to go backstage to cry.” “Cry? He’s a clown, Bill. Isn’t that part of the job?” “He told me. Said the kids keep pinning the tail on his face. It’s humiliating.” “Still, how do we know the cake isn’t rigged?” “The cake? Molly, what are you—?” “I read about this. Fancy parties. Trap cake. You slice it, alarm goes off, confetti cannons shoot netting. Happened in Belgium.” “That was a Bond film.” “Still! We can’t be sure. What if the birthday kid is a genius? Seven years old, builds laser grids in his sleep.” Bill squinted through the window. “He’s currently trying to eat glitter.” “Distraction technique. Classic misdirection.” “I think he’s just a kid with poor impulse control.” “So was Hannibal Lecter at nine.” Bill sighed. “Look, we go in when the clown sobs in the supply closet. We grab the presents, no names on tags—clean resale. Easy.” “What if they’ve got balloon animals guarding the loot?” “Balloon animals can’t guard anything unless they’re sword-wielding poodles.” “Never underestimate inflatable espionage.” “Molly, it’s a piñata full of glow sticks and plastic dinosaurs.” “You don’t know that. Could be diamonds. Kid’s rich—look at that robot butler pouring juice.” “That’s Greg. He’s a high school senior, minimum wage.” “Exactly. A spy.” Bill rubbed his temples. “Two minutes. Clown cries. We move. No souvenirs.” “But the tiny cowboy hat…” “No souvenirs.”
created_at
"2026-03-25T10:14:16.474Z"
3/25/2026, 10:14:16 AM
model
"qwen/qwen3-235b-a22b-07-25"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
70
moderation_latency
(null)
generation_time
5290
tokens_prompt
77
tokens_completion
498
native_tokens_prompt
72
native_tokens_completion
399
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
0.0000471
router
(null)
provider_responses
0
endpoint_id
"c9f927db-d559-49eb-9846-80862c5e3b4e"
id
"chatcmpl-ab11d6d3154bc720"
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-1774433656-YFm0k3tYMggSuOcIP8eM"
upstream_id
"chatcmpl-ab11d6d3154bc720"
total_cost
0.0000471
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
270
99.73%
Dialogue to Total Word Ratio
Ratio: 92.28%, Deviation: 2.28%
neededClean
false
wordsTotal
272
wordsDialogue
251
49.8652%