NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHYAVM3KTG9Q0QQVACFDAVNM
Run Details
Model:
Ministral 3 8B
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
3.8s
Cost:
$0.000115
Provider:
Mistral
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 500 word scene that has 30% 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
The neon glow of a flickering *Open* sign cast long shadows across the sidewalk as Molly adjusted the brim of her too-tight beanie, squinting at the three-story mansion ahead. "Alright, Bill," she muttered, "if we’re doing this, we need to move like we’re not doing this." Bill, perched on the fire escape like a disgruntled raccoon, adjusted his wire-rimmed glasses with a sigh. "I *told* you we should’ve gone for the toy store instead. At least there, the security’s just a guy named Gary who takes naps." "Gary’s got a point," Molly shot back, "but this place? It’s got *balloons*. And balloons mean presents. And presents mean *money*." She tapped her chin. "Also, the parents will be too busy pretending they’re having fun to notice us slip in like ghosts." Bill scoffed, swinging his legs over the railing. "Ghosts don’t steal from kids’ parties. Ghosts don’t get caught with glitter on their shoes." Molly grinned. "That’s why we’re not ghosts. We’re *professionals*." She pulled a crumpled map from her pocket, unfolding it with dramatic flair. "Phase one: Distraction. You’re on balloon duty." Bill groaned. "I *hate* balloons." "Good. That means you’ll be *excellent* at it." She pointed to the front yard, where a cluster of kids were already blowing up their faces like overinflated piñatas. "You pop a few, maybe knock over a piñata, get the little monsters screaming. I’ll slip in through the back patio door—" "—which is *locked*," Bill interrupted. "—*currently*. But I’ve got a lockpick set that’ll have it open faster than a kid can say ‘uncle.’" She wiggled her fingers. "And if that fails, I’ve got a spare key shaped like a dinosaur. Kids love dinosaurs." Bill eyed her skeptically. "You’re not telling me you *stole* that dinosaur key from a kid’s birthday party last year." Molly gasped, clutching her chest. "First of all, it was *borrowed*. Second, that was *two years* ago. Third—" She held up a finger. "—if we’re caught, I’m blaming you for the balloon plan." Bill sighed, rubbing his temples. "Fine. But if I get stuck popping balloons, I’m naming the first present I steal after you." Molly smirked. "Deal. And if I get caught, I’m naming *you* after the piñata." They exchanged a look—equal parts nervous and exhilarated—as the distant sound of laughter and squeals drifted from the house. Outside, the streetlights flickered, casting everything in a warm, golden haze. Somewhere inside, a kid was already unwrapping a shiny new toy, oblivious to the fact that in less than an hour, it would be in their pockets. Bill adjusted his gloves. "Alright. Let’s do this before I change my mind." Molly grinned, tucking her lockpicks into her sleeve. "Too late. We’re already in."
created_at
"2026-02-20T20:11:31.837Z"
2/20/2026, 8:11:31 PM
model
"mistralai/ministral-8b-2512"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
189
moderation_latency
(null)
generation_time
2603
tokens_prompt
77
tokens_completion
732
native_tokens_prompt
69
native_tokens_completion
699
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
0.0001152
router
(null)
provider_responses
0
endpoint_id
"c85d26e1-1e0c-449b-9775-2afce7ae510b"
id
"da2981ae9b6b41e7a7ac3b843760cfa3"
is_byok
false
latency
189
model_permaslug
"mistralai/ministral-8b-2512"
provider_name
"Mistral"
status
200
api_type
"completions"
id
"gen-1771618291-5C4tSgjhPzvm9Na0FV9q"
upstream_id
"da2981ae9b6b41e7a7ac3b843760cfa3"
total_cost
0.0001152
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Mistral"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
459
0.00%
Dialogue to Total Word Ratio
Ratio: 53.81%, Deviation: 23.81%
neededClean
false
wordsTotal
459
wordsDialogue
247
0.0000%