NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM09H2DFTVM3RRARK3EPQK3W
Run Details
Model:
Mistral Small 4
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
2.9s
Cost:
$0.000190
Provider:
Mistral
Evaluation summary
Score
Evaluator
Details Link
20.19%
Matches word count
(details)
59.79%
Dialogue to Total Word Ratio
(details)
39.9907%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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 neon glow of the "Happy Birthday, Timmy!" banner flickered weakly against the warehouses’ grimy windows as Molly adjusted her tool belt, her fingers already twitching with impatience. Bill, on the other hand, was busy second-guessing their entire plan by kneeling to test the lock on the back door with a pair of bent hairpins. "This is the third time you’ve done this," Molly muttered, her voice edged with the kind of dry humor people reserve for when they’re seconds away from committing grand larceny. Bill shot her a look, his face illuminated by the glow of his tiny penlight. The lock clicked. "Just making sure," he said, shoving the door open before Molly could groan. They slipped inside, the sudden quiet of an empty party room swallowing their footsteps. Balloons sagged near the ceiling, and the remains of a half-eaten cake sat on a table, surrounded by abandoned party hats and a single, gaudy clown figurine that seemed to sneer at them. Molly’s eyes locked onto a pile of presents stashed in the corner. Bill, meanwhile, was already eyeing the clown. "What?" Molly hissed when she caught him staring. "Think it’s real?" Bill whispered, poking the clown’s floppy hat. She exhaled sharply through her nose. "If it is, I don’t want to know." They had a job to do.
created_at
"2026-03-18T10:58:09.98Z"
3/18/2026, 10:58:09 AM
model
"mistralai/mistral-small-2603"
1/1/2603, 12:00:00 AM
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
363
moderation_latency
(null)
generation_time
2803
tokens_prompt
77
tokens_completion
336
native_tokens_prompt
81
native_tokens_completion
296
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
0.00018975
router
(null)
provider_responses
0
endpoint_id
"f5ed6283-4e17-457d-98cd-2ec0c318171d"
id
"ddb6af15d59b46988ace428c22562ab4"
is_byok
false
latency
363
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-1773831489-Z2Xa6qTbuMR59uGjJIRQ"
upstream_id
"ddb6af15d59b46988ace428c22562ab4"
total_cost
0.00018975
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Mistral"
Evaluation details
Result
Evaluator
Details
Meta Data
20.19%
Matches word count
n/a
neededClean
false
words
220
59.79%
Dialogue to Total Word Ratio
Ratio: 18.47%, Deviation: 8.47%
neededClean
false
wordsTotal
222
wordsDialogue
41
39.9907%