NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHXPK1XND05TX516QKAW61S6
Run Details
Model:
Mistral Medium 3.1
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
5.4s
Cost:
$0.000656
Provider:
Mistral
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.9995%
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
The rainbow balloons bobbed cheerfully in the breeze as Molly crouched behind a bush, adjusting her oversized clown wig. "You sure this is the right house? These kids look like they’ve got *way* too much glitter." Bill squinted at the bouncing castle in the backyard. "Positive. The invite said ‘Princess Jasmine’s Unicorn Extravaganza.’ That’s code for ‘rich parents who buy their kid a pony instead of love.’" Molly tugged at her polka-dot jumpsuit. "I still don’t get why we’re dressed as clowns. Couldn’t we just, I dunno, *walk in* like normal people?" "Normal people don’t steal from six-year-olds, Mol. And clowns? Invisible at kids’ parties. Trust me, no one looks twice at a guy with a red nose and a sad balloon animal." He held up a limp giraffe. "Exhibit A." She sighed. "Fine. But if I have to make one ‘honking’ noise, I’m taking the pony as my cut." A shriek of laughter erupted from the backyard. Molly’s eyes narrowed. "You hear that? That’s the sound of *unopened LEGO sets*." Bill grinned. "Phase one: blend in. Phase two: ‘accidentally’ knock over the gift table. Phase three—" "—profit." Molly cracked her knuckles. "Let’s go ruin some childhoods."
created_at
"2026-02-20T14:17:19.552Z"
2/20/2026, 2:17:19 PM
model
"mistralai/mistral-medium-3.1"
3/1/2001, 12:00:00 AM
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
238
moderation_latency
(null)
generation_time
5320
tokens_prompt
77
tokens_completion
320
native_tokens_prompt
69
native_tokens_completion
314
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
0.0006556
router
(null)
provider_responses
0
endpoint_id
"13c5002b-8c5b-490c-8a5f-52ec0242804f"
id
"9e52ada4f67b4cb8aad6b667d6e82790"
is_byok
false
latency
238
model_permaslug
"mistralai/mistral-medium-3.1"
3/1/2001, 12:00:00 AM
provider_name
"Mistral"
status
200
api_type
"completions"
id
"gen-1771597039-nHaiB894jw1sjgSZUHER"
upstream_id
"9e52ada4f67b4cb8aad6b667d6e82790"
total_cost
0.0006556
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Mistral"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
201
0.00%
Dialogue to Total Word Ratio
Ratio: 71.64%, Deviation: 21.64%
neededClean
false
wordsTotal
201
wordsDialogue
144
49.9995%