NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM08SYSWCMGF6883MA99E1PF
Run Details
Model:
Mistral Small 4
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
6.5s
Cost:
$0.000407
Provider:
Mistral
Evaluation summary
Score
Evaluator
Details Link
0.01%
Matches word count
(details)
78.32%
Dialogue to Total Word Ratio
(details)
39.1667%
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
Molly twirled a lockpick between her fingers as she crouched behind a dumpster, the stench of half-eaten cake and discarded balloons clinging to the air. Bill adjusted his clown wig for the tenth time, the frizzy red curls doing little to hide his scowl. "Alright, listen up," Molly said, her voice a whisper barely audible over the distant shriek of a toddler. "The cake’s already been cut, which means the parents are distracted. We’ve got twenty minutes tops before they remember to open presents." Bill groaned, shifting the oversized polka-dot bowtie around his neck. "Twenty minutes? With these shoes? I can’t run in these things. Last heist, I tripped over a balloon animal and face-planted into a pile of glitter." Molly shot him a look. "Please tell me you left the glitter out of this plan." "It’s part of the disguise!" Bill protested. "How else am I supposed to blend in at a children’s party?" "By not looking like you’re being held hostage by a rave and a balloon-fighting tournament?" Molly sighed. "Fine. You go in through the back door like we rehearsed. I’ll handle the presents from the inside. The clown costume’s a liability, Bill, not a tactical advantage." Bill clutched his chest in mock offense. "A liability? Molly, this costume is *art*. The parents won’t suspect a thing when I start juggling stolen toys. In fact, they’ll probably tip me." Molly pinched the bridge of her nose. "Just get in, grab the good stuff, and meet me at the van. And for the love of all things holy, don’t eat any of the cake." Bill adjusted his oversized gloves. "What if it’s my favorite flavor?" "It’s vanilla." "Exactly." Molly exhaled sharply, then nodded toward the house. "Go. Now." Bill waddled off, his giant shoes flapping with each step. Molly watched him go, then muttered under her breath, "I’m robbing a kindergarten party with a man who moves like a constipated penguin." A child’s laughter rang out from inside. Molly checked her watch, counting the seconds. Ten to go. She stepped out of the shadows, her own disguise—fluffy pink sweater and a too-wide grin—already making her blend in. As she crossed the lawn, a tiny hand yanked on her sleeve. "Are you the new helper?" a little girl asked, her pigtails bouncing. Molly hesitated. "Uh. Yes. The assistant clown." The girl beamed. "Cool! Can you blow up more balloons?" Molly past the sudden urge to toss the child over her shoulder like a sack of stolen loot. "Sure thing, kid. After I help with the presents." The girl skipped away, and Molly slipped inside, her fingers already itching for the first wrapped box. Somewhere in the house, Bill was probably demanding more cake. Molly just hoped he didn’t blow their whole operation over a dessert-based meltdown.
created_at
"2026-03-18T10:45:32.618Z"
3/18/2026, 10:45:32 AM
model
"mistralai/mistral-small-2603"
1/1/2603, 12:00:00 AM
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
394
moderation_latency
(null)
generation_time
6429
tokens_prompt
77
tokens_completion
729
native_tokens_prompt
81
native_tokens_completion
658
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
0.00040695
router
(null)
provider_responses
0
endpoint_id
"f5ed6283-4e17-457d-98cd-2ec0c318171d"
id
"74a2fb669781411583f7f3e62ef2f189"
is_byok
false
latency
394
model_permaslug
"mistralai/mistral-small-2603"
1/1/2603, 12:00:00 AM
provider_name
"Mistral"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1773830732-Wv7RCziFzobpC3VorPjU"
upstream_id
"74a2fb669781411583f7f3e62ef2f189"
total_cost
0.00040695
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Mistral"
Evaluation details
Result
Evaluator
Details
Meta Data
0.01%
Matches word count
n/a
neededClean
false
words
469
78.32%
Dialogue to Total Word Ratio
Ratio: 47.79%, Deviation: 2.21%
neededClean
false
wordsTotal
475
wordsDialogue
227
39.1667%