NC Bench
Overview
About
Tests
Models
Model
Claude 3.0 Sonnet
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-07-19-01J35SQ6PP0GYC90Z5TC565S9X
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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
"So, Molly, ya ready for the big score?" Bill asked, rubbing his hands together with a mischievous grin. Molly rolled her eyes. "Stealing presents from a bunch of snot-nosed kids? Really, Bill? That's low, even for us." "Ah, but that's where the beauty lies! Who'd suspect a pair of harmless, jolly thieves at a kiddie party?" "I dunno, Bill. Seems risky. What if we get caught? Do you really want to go down in history as the Grinch who stole Christmas?" Bill waved his hand dismissively. "Psh, like that green furball could hold a candle to us! Besides, we'll be in and out before anyone notices. Just think of all the loot – video games, dolls, RC cars. It'll be like hitting the jackpot at Toys 'R' Us!" Molly pondered for a moment, then a devious smile crept across her face. "Alright, you've twisted my arm. But if we're gonna do this, we're gonna do it right. I'm talking full disguises – Santa suits, fake beards, the works." "Now you're talking!" Bill exclaimed, rubbing his hands together eagerly. "Operation Heist-mas is a go!"
id
"gen-NFL6xOybZ6k9AnWzFInr9Dk5j3gP"
total_cost
0.004785
upstream_id
"msg_01RT7bCtAaxsiqQXYhmamNUE"
created_at
"2024-07-19T15:28:18.48743+00:00"
Friday, July 19th, 2024 at 3:28:18 PM GMT+00:00
model
"anthropic/claude-3-sonnet:beta"
app_id
182717
streamed
false
cancelled
false
provider_name
"Anthropic"
latency
7124
moderation_latency
(null)
generation_time
0
finish_reason
"end_turn"
tokens_prompt
68
tokens_completion
265
native_tokens_prompt
75
native_tokens_completion
304
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.004785
Evaluation details
Result
Evaluator
Details
Meta Data
43.3783%
Matches word count
n/a
neededClean
false
words
183
19.4542%
Dialogue to Total Word Ratio
Ratio: 78.69%, Deviation: 11.31%
neededClean
false
wordsTotal
183
wordsDialogue
144
31.4163%