NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHYB38R63QSZMV5SFDHNXYZA
Run Details
Model:
Ministral 3 3B
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
2.7s
Cost:
$0.000083
Provider:
Mistral
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0001%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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
**Title: *The Great Toy Heist*** The back door of the McGinty mansion creaked open with a sound like a sleepy toddler’s sneeze. Molly adjusted her wire-rimmed glasses and peered through the peephole, squinting at the glittering chaos inside. "Bill, you’re *not* gonna believe this." Bill wiped his brow with a napkin, already eyeing the glittery tablecloth. "Molly, I *do* believe it’s the most expensive party I’ve ever seen. And I *do* believe we’re about to get arrested for trespassing." Molly rolled her eyes. "Relax, Bill. We’re not here to arrest anyone. We’re here to arrest *the presents*." A giggling voice piped up from inside. "Mom! Dad! The presents are *everywhere*!" Bill groaned. "Oh, come *on*. You’re gonna have to be more specific." Molly grinned, already pulling out her lockpick. "Fine. The ones under the tree. The ones wrapped in *actual* paper, not those plastic ones. And the ones that aren’t already in the hands of a very confused 5-year-old." A shadow moved in the doorway. "You two are *so* not invited." Bill spun around, nearly dropping his flashlight. "Oh, *shit*—it’s the security guard." Molly’s eyes widened. "Bill, he’s *not* a security guard. He’s a *very* confused security guard." The man—who looked like he’d been dragged from a nap—stumbled forward, holding a half-eaten cupcake. "Uh… hi? You two look like you’re up to no good." Bill adjusted his sunglasses. "We’re just… uh… *collecting*." The man blinked. "Collecting what?" Molly stepped forward, holding up a glittery paper bag. "These. The *best* presents. The ones that don’t belong to anyone yet." The man’s eyes darted to the door. "You’re not supposed to be here." Bill sighed. "We know. But we *need* the presents. For… uh… *artistic purposes*." The man hesitated, then pointed at the tree. "The presents are under the tree." Molly nodded. "Perfect. Bill, you grab the ones wrapped in *real* paper. I’ll handle the ones with *extra* glitter." Bill groaned. "Molly, you’re gonna get us both arrested." She winked. "Nah. I’ve got a *plan*." A loud *BANG* echoed from inside. The man jumped. "Oh no. Someone’s *not* invited." Molly grinned. "That’s our cue." Bill grabbed his bag, already sprinting toward the tree. "Go! Go! Go!" Molly followed, dodging a confused toddler who was trying to hug her. "Watch where you’re *going*, kid!" Inside, chaos erupted. A security guard in a Santa hat tried to stop them. "You can’t take the presents!" Bill dodged him, grabbing a wrapped gift. "We *can*. And we *will*." Molly snatched another, slipping out the window just as the guard called after them. "You two are *so* not invited!" Bill grinned, already counting his loot. "Nah. We’re *invited*. To the next level." Molly adjusted her glasses. "Level up. *Toy level.*" And with that, the greatest heist in history began.
created_at
"2026-02-20T20:15:42.351Z"
2/20/2026, 8:15:42 PM
model
"mistralai/ministral-3b-2512"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
201
moderation_latency
(null)
generation_time
2375
tokens_prompt
77
tokens_completion
781
native_tokens_prompt
69
native_tokens_completion
763
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
0.0000832
router
(null)
provider_responses
0
endpoint_id
"b9f6828f-aef9-498a-91f5-6816fbf72420"
id
"abf5fd6ff49a41fbb596a9afbd9227b0"
is_byok
false
latency
201
model_permaslug
"mistralai/ministral-3b-2512"
provider_name
"Mistral"
status
200
api_type
"completions"
id
"gen-1771618542-gzjBiga44cRGC3hVz04L"
upstream_id
"abf5fd6ff49a41fbb596a9afbd9227b0"
total_cost
0.0000832
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
466
0.00%
Dialogue to Total Word Ratio
Ratio: 50.64%, Deviation: 19.36%
neededClean
false
wordsTotal
466
wordsDialogue
236
0.0001%