NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY9MZGVQ98REPYRSSTR21A4
Run Details
Model:
Ministral 3 8B
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
3.2s
Cost:
$0.000103
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 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
The neon glow of a flickering "Happy Birthday, Timmy!" balloon cast long shadows across the living room floor as Molly adjusted the strap of her satchel, her breath fogging the glass of the sliding glass door. "Alright, Bill," she whispered, nudging him with her elbow. "We’re in. But we’ve got to move fast—Timmy’s parents are *way* too into this ‘open bar’ thing. You ever see a grown-up chug a cupcake?" Bill, already crouched by the front door, grinned around a mouthful of stolen candy canes. "I’ve seen it. It’s like watching a goldfish try to do backflips. And hey, if we’re stealing presents, we might as well make it *festive*. Got a plan for the security?" Molly rolled her eyes, pulling a tiny remote from her pocket. "This? It’s a *remote*. For the parents’ phones. One press, and they’re all asleep in the living room, snoring like a chorus of drunk penguins. But we’ve got to time it right—Timmy’s got a ‘surprise’ speech at, like, 8:30. So we hit the house by 8:15 sharp, or we’re stealing from a room full of crying kids and confused adults." Bill adjusted his wire-frame glasses, squinting at his watch. "You sure about this? Last time we did a kids’ party, we ended up with a piñata full of *glow sticks* and a toddler who swore we were Santa’s ‘backup elves.’" Molly smirked, tossing him a small, velvet-lined pouch. "Relax, Bill. This time, we’re not stealing *from* the kids—we’re stealing *for* them. Imagine: no more ‘one present per kid’ rules. No more ‘you can’t have that one because it’s *too nice*.’ We’re giving them the *whole* treasure chest. Literally. That’s the one under the tree with the ‘DO NOT OPEN’ sign. I saw it." Bill groaned, clutching his chest. "You’re evil. But fine. I’ll handle the ‘distraction’—I’ll ‘accidentally’ knock over the cake stand and make it look like I’m tripping over my own feet. You handle the presents." Molly grinned. "Deal. But if we get caught, I’m blaming you for the ‘accidental’ cake stand. And also for the fact that you once tried to steal a teddy bear and ended up with a *stuffed raccoon* instead." Bill laughed, already slipping out the door. "Worth it. Meet you by the tree in three minutes. And Molly?" She turned, one eyebrow raised. "Bring the duct tape. Just in case." Molly sighed, pulling out a roll from her pocket. "Always."
created_at
"2026-02-20T19:50:25.574Z"
2/20/2026, 7:50:25 PM
model
"mistralai/ministral-8b-2512"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
192
moderation_latency
(null)
generation_time
2886
tokens_prompt
77
tokens_completion
631
native_tokens_prompt
69
native_tokens_completion
621
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
64
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
0.0001035
router
(null)
provider_responses
0
endpoint_id
"c85d26e1-1e0c-449b-9775-2afce7ae510b"
id
"5b2b84fedad442b6ac4a6fce8a4f8924"
is_byok
false
latency
192
model_permaslug
"mistralai/ministral-8b-2512"
provider_name
"Mistral"
status
200
api_type
"completions"
id
"gen-1771617025-rEVd1ZzD0RQ5S1olGWSm"
upstream_id
"5b2b84fedad442b6ac4a6fce8a4f8924"
total_cost
0.0001035
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
409
0.00%
Dialogue to Total Word Ratio
Ratio: 72.37%, Deviation: 22.37%
neededClean
false
wordsTotal
409
wordsDialogue
296
0.0000%