NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHXPXHB8AHBVER0NDKBRTAZE
Run Details
Model:
Mistral Small 3.2 24B
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
6.3s
Cost:
$0.000239
Provider:
Mistral
Evaluation summary
Score
Evaluator
Details Link
60.28%
Matches word count
(details)
0.09%
Dialogue to Total Word Ratio
(details)
30.1819%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 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 the oversized sunglasses perched on her nose and squinted at the brightly colored banner fluttering in the breeze. "You sure this is the right place, Bill? It looks like a rainbow exploded on a cupcake." Bill, balancing a stack of stolen party supplies in his arms, grinned. "Positive. The kid’s parents are loaded. Presents will be top-tier. Maybe even a drone." Molly snorted. "A drone? At a five-year-old’s birthday? What’s next, a timeshare?" "Hey, rich people are weird. Last year, I hit a bar mitzvah with a solid gold yarmulke. Solid gold, Molly." "Focus. We need a plan. How do we get in without looking suspicious?" Bill tapped his chin. "Easy. We’re the clowns." Molly blinked. "We’re what?" "The clowns. The original ones canceled. I heard the mom on the phone. She’s desperate." Molly groaned. "Bill, we can’t clown. We’re terrible at clowning." "Nonsense. We’ve got the outfits. I’ve got a nose honker." Molly crossed her arms. "A what?" Bill pulled a red rubber bulb from his pocket and squeezed it. A loud, obnoxious honk filled the air. Molly winced. "That’s not a nose, that’s a foghorn." "Details. We’ll wing it. Balloons, silly walks, the usual." Molly sighed. "Fine. But if I have to juggle, I’m quitting." Bill clapped her on the back. "You’ll be great. Just remember—big smiles, no murder." Molly adjusted her oversized shoes. "No promises." They approached the house, where a frazzled woman in a floral dress waved them inside. "Oh, thank goodness you’re here! The kids are already losing it." Bill bowed dramatically. "Fear not, madam. We are the Clowns of Chaos." The woman blinked. "That’s… not the name on the contract." Molly elbowed Bill. "He’s poetic. Let’s go." Inside, a dozen children screamed and threw confetti. Molly’s stomach twisted. "This is a war zone." Bill grinned. "Perfect. Distraction is key. I’ll do the balloon animals. You… uh… do the…" Molly glared. "The what?" "Uh… the… honking?" Molly grabbed the foghorn and blasted it. The kids shrieked with delight. "Fine. But if any of them cry, I’m blaming you." Bill saluted and began twisting balloons into lopsided giraffes. Molly wandered the room, honking occasionally, her eyes scanning the gift table. "Jackpot. There’s a Lego set the size of a small car." Bill joined her, whispering. "And look—over there. A tablet. Unlocked." Molly’s eyes gleamed. "We’re stealing a tablet at a kid’s party. We’re monsters." Bill shrugged. "Monsters with a plan. Now, how do we get the goods without getting caught?" Molly honked the foghorn again, then pointed. "See that kid in the dinosaur costume? He’s not even looking. I’ll swap the presents with the empty boxes we brought." Bill nodded. "And I’ll ‘accidentally’ knock over the cake. Chaos covers everything." Molly grinned. "You’re a genius." Bill winked. "I know. Now let’s get rich." They high-fived as the party descended into glorious, honking madness.
created_at
"2026-02-20T14:23:03.027Z"
2/20/2026, 2:23:03 PM
model
"mistralai/mistral-small-3.2-24b-instruct-2506"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
196
moderation_latency
(null)
generation_time
6116
tokens_prompt
77
tokens_completion
818
native_tokens_prompt
69
native_tokens_completion
773
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
0.0002388
router
(null)
provider_responses
0
endpoint_id
"23fd6b91-74d8-45fc-ac00-cf141ec7f4a5"
id
"89cf414189cc42bd8795888f0ad3e722"
is_byok
false
latency
196
model_permaslug
"mistralai/mistral-small-3.2-24b-instruct-2506"
provider_name
"Mistral"
status
200
api_type
"completions"
id
"gen-1771597383-oGWdme6ndKlR4u6UhzT0"
upstream_id
"89cf414189cc42bd8795888f0ad3e722"
total_cost
0.0002388
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Mistral"
Evaluation details
Result
Evaluator
Details
Meta Data
60.28%
Matches word count
n/a
neededClean
false
words
485
0.09%
Dialogue to Total Word Ratio
Ratio: 61.86%, Deviation: 11.86%
neededClean
false
wordsTotal
485
wordsDialogue
300
30.1819%