NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHXPC4A2TXWVYYJFQ4NSF83Y
Run Details
Model:
Mistral Large 3
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
16.8s
Cost:
$0.001084
Provider:
Mistral
Evaluation summary
Score
Evaluator
Details Link
99.74%
Matches word count
(details)
9.81%
Dialogue to Total Word Ratio
(details)
54.7766%
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
Molly adjusted the glittery unicorn horn strapped to her forehead and squinted at the clipboard in her hands. "Alright, Bill, let’s go over this again. The party starts at three, which means we’ve got exactly forty-seven minutes to get in, grab the loot, and get out before the sugar crash hits and those kids turn into tiny, screaming banshees." Bill, dressed in a suspiciously clean clown costume, frowned at his oversized shoes. "I still don’t see why I have to be the clown. Clowns are creepy. Have you ever met a clown who wasn’t at least *a little* unhinged?" "Because, my dear Bill, you’re the one with the face for it. Also, you’re the only one who can pull off that nose without looking like you’re smuggling a cherry under there." Molly tapped the clipboard. "Now, focus. The parents are distracted by the magician, the kids are hyped up on cake, and the presents are all piled up by the gift table like a buffet of unguarded treasure." Bill sighed, adjusting his rainbow wig. "Fine. But if one of those kids tries to hug me, I’m out. I didn’t sign up for emotional labor." Molly rolled her eyes. "You’re a thief, not a therapist. Just smile, hand out balloons, and for the love of all things criminal, *do not* make eye contact with the parents. We don’t need some overprotective dad remembering your face when the cops come knocking." Bill pulled a balloon animal out of his pocket—a sad, lopsided giraffe. "What’s this for?" "Distraction. If anyone gets suspicious, you start twisting balloons into questionable shapes and suddenly no one’s looking at me slipping the presents into the giant piñata I brought." Bill held up the giraffe. "This looks like it lost a fight with a lawnmower." "Exactly. It’s so bad it’s *memorable*. They’ll be talking about your ‘art’ for years, not the fact that half the gifts vanished." Molly checked her watch. "Alright, we’ve got ten minutes. You go in first, act like you’re part of the entertainment. I’ll follow in five, dressed as a very concerned aunt who ‘just stopped by to drop off a gift.’" Bill groaned. "I hate this plan. What if the magician recognizes me from that time I tried to pick his pocket at the county fair?" Molly patted his shoulder. "Then you improvise. Tell him you’re his biggest fan. Tell him you’re his long-lost twin. Tell him you’re a time traveler here to prevent the invention of balloon animals. I don’t care, just keep him busy." Bill took a deep breath. "Fine. But if this goes sideways, I’m blaming you." Molly grinned. "That’s the spirit. Now go be terrifyingly cheerful." Bill waddled toward the party, his oversized shoes squeaking with every step. Molly counted to thirty before adjusting her cardigan and grabbing the piñata—a suspiciously lumpy, gift-shaped piñata. She muttered under her breath, "Alright, kids. Let’s see what Santa *really* brought you this year."
created_at
"2026-02-20T14:13:32.619Z"
2/20/2026, 2:13:32 PM
model
"mistralai/mistral-large-2512"
1/1/2512, 12:00:00 AM
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2576
moderation_latency
(null)
generation_time
16754
tokens_prompt
77
tokens_completion
764
native_tokens_prompt
69
native_tokens_completion
700
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
0.0010845
router
(null)
provider_responses
0
endpoint_id
"fa718841-07ea-4b7e-b8a3-6069545c7e6a"
id
"42a8f88a9a4648d1b75abdc28bc56fb5"
is_byok
false
latency
2576
model_permaslug
"mistralai/mistral-large-2512"
1/1/2512, 12:00:00 AM
provider_name
"Mistral"
status
200
api_type
"completions"
id
"gen-1771596812-n3JwpvyT4TobzI248qIO"
upstream_id
"42a8f88a9a4648d1b75abdc28bc56fb5"
total_cost
0.0010845
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Mistral"
Evaluation details
Result
Evaluator
Details
Meta Data
99.74%
Matches word count
n/a
neededClean
false
words
496
9.81%
Dialogue to Total Word Ratio
Ratio: 76.81%, Deviation: 6.81%
neededClean
false
wordsTotal
496
wordsDialogue
381
54.7766%