NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM08VZNAHQM503QTD9HPSQPD
Run Details
Model:
Mistral Small 4
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
3.2s
Cost:
$0.000197
Provider:
Mistral
Evaluation summary
Score
Evaluator
Details Link
99.98%
Matches word count
(details)
60.55%
Dialogue to Total Word Ratio
(details)
80.2689%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 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 adjusted her candy-striped beanie and eyed the bouncy castle in the backyard with a mix of admiration and greed. "You sure this place is packed with toys?" she whispered to Bill, who was crouched beside her, crunching on stale pretzels from a stolen bag. Bill wiped away crumbs and nodded. "Sign says 'kids’ birthday bonanza—limited edition gaming console, robotic hamsters, the works.'" Molly grinned. "Limited edition means undefended." She patted her utility belt, which held more lock picks than Band-Aids. "We grab the good stuff, drop it off at the fencing pawn shop by midnight, and—" "By *midnight*?" Bill cut in. "That’s when their security guard does his rounds! The dude’s ex-military. Kid once karate-chopped a piñata into confetti." Molly flicked his forehead. "Hence the distraction plan." She pulled out a whoopee cushion with a grin. "We *gently* place this in the clown’s microphone. The room erupts in laughter. Security thinks it’s all fun and games. Meanwhile, we’re in the gift room like ninjas with hangovers." Bill scratched his ear. "What if the clown detects the whoopee cushion?" Molly tossed him the device. "First rule of heists: If you’re second-guessing the whoopee cushion, you’ve already lost."
created_at
"2026-03-18T10:46:39.035Z"
3/18/2026, 10:46:39 AM
model
"mistralai/mistral-small-2603"
1/1/2603, 12:00:00 AM
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
369
moderation_latency
(null)
generation_time
3051
tokens_prompt
77
tokens_completion
327
native_tokens_prompt
81
native_tokens_completion
311
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
16
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
0.00019659
router
(null)
provider_responses
0
endpoint_id
"f5ed6283-4e17-457d-98cd-2ec0c318171d"
id
"4b506c4a5020496abe4d6adb67311bd6"
is_byok
false
latency
369
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-1773830799-kKmW4zELbhzSFppINPpl"
upstream_id
"4b506c4a5020496abe4d6adb67311bd6"
total_cost
0.00019659
cache_discount
0.00000216
upstream_inference_cost
0
provider_name
"Mistral"
Evaluation details
Result
Evaluator
Details
Meta Data
99.98%
Matches word count
n/a
neededClean
false
words
198
60.55%
Dialogue to Total Word Ratio
Ratio: 58.42%, Deviation: 8.42%
neededClean
false
wordsTotal
202
wordsDialogue
118
80.2689%