NC Bench
Overview
About
Tests
Models
Model
Claude 3.5 Sonnet
Test
Dialogue tags
Scenario
Write 200 words with 10% dialogue
Run ID
2024-07-19-01J35T8J9W9KQD706ZWC3ST1E9
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
Molly hunched over the kitchen table, her eyes darting between a crudely drawn map and a half-eaten slice of pizza. Bill paced behind her, nervously twirling a party blower between his fingers. "You sure about this, Mol?" Bill asked, his voice cracking. Molly rolled her eyes. "It's foolproof. We go in dressed as clowns, blend in with the entertainment, and snatch the loot while everyone's distracted by the magician." Bill's face contorted with worry. "But... it's a kid's party. Doesn't that seem a bit... y'know, wrong?" "Wrong? Bill, we're thieves. It's what we do," Molly retorted, circling a spot on the map with a greasy finger. The room fell silent as Bill contemplated their plan. Molly stood up, stretching her arms above her head. She walked to the fridge, pulled out two sodas, and tossed one to Bill. "Think of it this way," Molly said, cracking open her can. "We're teaching those little brats a valuable life lesson." Bill caught the soda, fumbling it slightly. "And what's that?" Molly grinned mischievously. "Trust no one, especially not clowns."
id
"gen-SdMRReGPzXyA4mHJIWdOgEdBq0q3"
total_cost
0.00477
upstream_id
"msg_01X7DQ1rVU3RT25D3ohusxLy"
created_at
"2024-07-19T15:37:53.591528+00:00"
Friday, July 19th, 2024 at 3:37:53 PM GMT+00:00
model
"anthropic/claude-3.5-sonnet:beta"
app_id
182717
streamed
false
cancelled
false
provider_name
"Anthropic"
latency
13442
moderation_latency
(null)
generation_time
0
finish_reason
"end_turn"
tokens_prompt
68
tokens_completion
261
native_tokens_prompt
75
native_tokens_completion
303
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.00477
Evaluation details
Result
Evaluator
Details
Meta Data
9.6081%
Matches word count
n/a
neededClean
false
words
178
0.0000%
Dialogue to Total Word Ratio
Ratio: 39.89%, Deviation: 29.89%
neededClean
false
wordsTotal
178
wordsDialogue
71
4.8041%