NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-07-01M4ATAMXVA9FTRAFZRMVXNPN7
Run Details
Model:
Mistral Large 4.0 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
6m 51s
Cost:
$0.065375
Provider:
Mistral
Evaluation summary
Score
Evaluator
Details Link
43.38%
Matches word count
(details)
99.12%
Dialogue to Total Word Ratio
(details)
71.2504%
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
The warehouse smelled of coffee and ozone. Molly spread blueprints on a folding table, marker leaking onto her thumb. Bill wore a yellow polo, bouncing a rubber ball. Outside, a siren wailed, but neither looked up. The fluorescent lights hummed overhead. "We're actually doing this?" Molly asked, her voice tight. "We're robbing a six-year-old's birthday party in broad daylight?" "It's a target rich environment," Bill said, catching the ball. "Twenty kids, each with five presents. That's a hundred gifts wrapped in shiny paper by the cake table." "Bill, they're children." "Affluent children. Timmy's dad drives a Tesla. Timmy doesn't need a Lego Death Star. We're building character." Molly pinched her nose and stood up, chair scraping against the concrete floor. "We're stealing from kids, Bill. Actual kids with sticky fingers and juice boxes." "And we'll be very quiet," Bill said, snapping a clown nose onto his face. "I'm Bozo. You're Giggles. We create chaos, grab the loot, vanish like ninjas." "I'm not wearing a dress with polka dots." "You wear what I say, please. I'm the mastermind." Molly spun around, arms crossed. "You said natural history museum. Dinosaur bones. Classy." "Plans change. Museums have lasers. This place has a bounce house. What security is that?" "The kind with parents, Bill. Twenty sets of parents. Mothers with diaper bags and fathers with permits. And probably off-duty cops." Bill grinned, eyes gleaming. "Details. We hit during cake. I use a giant whoopee cushion. You grab presents. Escape through the ball pit." "The ball pit is three feet deep," Molly said, sitting back down. "Filled with plastic balls and lost shoes. We'll drown in there." "Exactly! No one suspects the ball pit. It's the perfect criminal thoroughfare. Unpredictable. Chaotic." Molly sighed and picked up the marker. "How do we carry out forty pounds of presents without looking suspicious?" "We rent a van. Disguised as a party truck. I bought magnetic signs from a guy named Steve." "Bill, you didn't." "I did. 'Penny's Party Palace.' We park by the loading dock and blend in with parents." Molly stared at him, marker hovering. She looked at his ridiculous nose, his manic eyes, then at the presents on her map, representing pure joy. "You planned this?" "Meticulously. Pizza at noon, cake at twelve-fifteen. We hit right at twelve-oh-five when the sugar rush starts." "And if we get caught?" Bill adjusted his clown nose, straightening it. "We say we're undercover. Investigating a birthday party crime ring." "There is actually no birthday party crime ring." "There will be. Think of the haul. The PS5s. The sneakers. The gift cards. Could afford that boat." Molly looked at the blueprints, then at Bill's nose. She thought about her eviction notice. A wicked smile crept across her face. "Fine. But if a kid cries, you buy the replacement present out of your share." "Deal. Now practice your laugh. It needs to be really unsettling. More clown-like." "My laugh is normal." "It sounds like a strangled seagull." "That's character." Bill stood up, knocking over the crates. "Then let's get to work. Bozo and Giggles have a party to crash."
created_at
"2026-10-07T09:15:18.852Z"
10/7/2026, 9:15:18 AM
model
"mistralai/mistral-large-4-0-20261006"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2426
moderation_latency
(null)
generation_time
410573
tokens_prompt
75
tokens_completion
29181
native_tokens_prompt
67
native_tokens_completion
31258
native_tokens_completion_images
(null)
native_tokens_reasoning
28368
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
service_tier
"default"
usage
0.06537478
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791364518-MdWKgmQEk3g7lrbTp918"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791364518-KwNQajB76RLrDZ6Nq2qX"
upstream_id
"ec2170ddd63e40d1b90c5e61cc18cf18"
provider_responses
0
endpoint_id
"19b3b5d7-23f6-4b63-8109-64a6cc37d996"
id
"ec2170ddd63e40d1b90c5e61cc18cf18"
is_byok
false
latency
2385
model_permaslug
"mistralai/mistral-large-4-0-20261006"
provider_name
"Mistral"
status
200
total_cost
0.06537478
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Mistral"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
43.38%
Matches word count
n/a
neededClean
false
words
517
99.12%
Dialogue to Total Word Ratio
Ratio: 70.42%, Deviation: 0.42%
neededClean
false
wordsTotal
524
wordsDialogue
369
71.2504%