NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY8JNF661Y89R7QWENFGH7K
Run Details
Model:
Ministral 3 3B
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
2.3s
Cost:
$0.000065
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 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 neon glow of the party’s disco ball flickered against Molly’s sneakers as she adjusted her wire-rimmed glasses. Bill leaned against the fence, arms crossed, watching the chaos inside. “You’re sure this is the right place?” Bill asked, voice dripping with skepticism. “Positive,” Molly said, pulling out her phone. “The security camera’s got a dead pixel right over the cake table. Perfect.” Bill groaned. “You’re telling me we’re about to rob a *children’s party*?” “Exactly,” Molly said, tossing him a bag of stolen candy. “And if we’re caught, we’ll just say we were *supposed* to be there for the ‘treasure hunt.’” “You’re insane.” “And you’re the one who suggested we do this,” Molly shot back. “You’re the one who’s always got a panic attack at the sight of a kid’s laughter.” Bill sighed, rubbing his temples. “Fine. But if we get caught, I’m blaming you.” “Deal.” Molly grinned, already unzipping her backpack. “First rule—no screaming. Second rule—if we see a parent, we *pretend* to be lost cousins.” Bill scoffed. “You’re a terrible liar.” “Nah,” Molly said, ducking through the gap in the fence. “I’m *perfect* at this.” Inside, the party was a symphony of giggles and clinking glasses. A group of kids in glow sticks circled a giant box labeled *OPEN ME!* while their parents milled about, oblivious. “Okay,” Bill whispered, eyeing the nearest parent. “What’s the play?” “Distract them,” Molly said, grabbing a nearby balloon. “I’ll pop it, and you fake a coughing fit.” Bill groaned. “You’re *terrible* at this.” “Shut up,” Molly said, hurling the balloon into the air. It burst with a dramatic *POP*, sending a cloud of confetti into the air. A woman turned, squinting. “Oh! My *balloon*!” Bill immediately doubled over, coughing into his fist. “You’re *so* bad at this,” Molly muttered, already slipping into the crowd. “Just wait till we get the presents,” Bill called after her. “Then we’ll see who’s the real thief.” Molly didn’t answer. She was already stealing a stuffed animal from a kid’s hands—*and* stealing his attention. “Hey! That’s mine!” the boy wailed. Bill, still coughing, just nodded and whispered, “We’re good.” Molly grinned. “We’re *so* good.”
created_at
"2026-02-20T19:31:41.17Z"
2/20/2026, 7:31:41 PM
model
"mistralai/ministral-3b-2512"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
305
moderation_latency
(null)
generation_time
1983
tokens_prompt
77
tokens_completion
663
native_tokens_prompt
69
native_tokens_completion
585
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
0.0000654
router
(null)
provider_responses
0
endpoint_id
"b9f6828f-aef9-498a-91f5-6816fbf72420"
id
"7665fc318ac74d8eb4372380aef5244b"
is_byok
false
latency
305
model_permaslug
"mistralai/ministral-3b-2512"
provider_name
"Mistral"
status
200
api_type
"completions"
id
"gen-1771615901-Q6RbeaFwNB2BmqrsAgdr"
upstream_id
"7665fc318ac74d8eb4372380aef5244b"
total_cost
0.0000654
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
365
0.00%
Dialogue to Total Word Ratio
Ratio: 44.66%, Deviation: 25.34%
neededClean
false
wordsTotal
365
wordsDialogue
163
0.0000%