NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHXPJZDQ18ENZZBGYKQ85FG2
Run Details
Model:
Mistral Medium 3.1
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
9.2s
Cost:
$0.000650
Provider:
Mistral
Evaluation summary
Score
Evaluator
Details Link
99.74%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.8722%
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 against the ceiling of the community center like overcaffeinated jellyfish. Molly adjusted her oversized clown bowtie and squinted at the "NO ADULTS WITHOUT A CHILD" sign. "You realize we’re about to commit felony-level party crashing, right?" Bill wiped glitter off his fake mustache. "Relax. We’re not crashing—we’re *enhancing*. Those kids don’t *need* thirty-seven Nerf guns. We’re redistributing joy. Robin Hood-style, but with more cake theft." Molly side-eyed the piñata shaped like a unicorn. "You ever actually *met* a seven-year-old? They’ve got the surveillance skills of a CIA operative and the moral flexibility of a pirate. One wrong move, and we’re getting tackled by a swarm of sugar-high vigilantes." Bill grinned, patting his fanny pack—stuffed with lockpicks and fruit snacks for "camouflage." "That’s why we go in as *entertainment*. You do the balloon animals, I’ll ‘accidentally’ knock over the gift table. Classic distraction." Molly sighed. "Fine. But if we get taken down by a kid in a Spider-Man costume, I’m blaming your lack of a backup plan." Bill tossed her a party hat. "Backup plan’s simple: if things go south, we start a conga line. No one questions the conga line."
created_at
"2026-02-20T14:17:16.993Z"
2/20/2026, 2:17:16 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
298
moderation_latency
(null)
generation_time
9113
tokens_prompt
77
tokens_completion
328
native_tokens_prompt
69
native_tokens_completion
311
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
48
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
0.0006496
router
(null)
provider_responses
0
endpoint_id
"13c5002b-8c5b-490c-8a5f-52ec0242804f"
id
"ae76d1de06384746a7523092a6af6884"
is_byok
false
latency
298
model_permaslug
"mistralai/mistral-medium-3.1"
3/1/2001, 12:00:00 AM
provider_name
"Mistral"
status
200
api_type
"completions"
id
"gen-1771597037-CSvDpTp8kvh1TyDZ2eV4"
upstream_id
"ae76d1de06384746a7523092a6af6884"
total_cost
0.0006496
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Mistral"
Evaluation details
Result
Evaluator
Details
Meta Data
99.74%
Matches word count
n/a
neededClean
false
words
204
0.00%
Dialogue to Total Word Ratio
Ratio: 69.61%, Deviation: 19.61%
neededClean
false
wordsTotal
204
wordsDialogue
142
49.8722%