NC Bench
Overview
About
Tests
Models
Model
Claude 3.5 Sonnet
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-07-19-01J35T9K69YJ8BKQGNPP4WT954
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
"Molly, you can't be serious," Bill whispered, his eyes wide with disbelief. "Dead serious, Bill. Think about it - a goldmine of gifts, all wrapped up and ready to go." "But it's a children's party! We can't steal from kids." "Oh, come on. These are rich brats. They've got more toys than they know what to do with." "Still, it feels wrong. What about Santa Claus and all that?" "Santa's not real, genius. Besides, we'll be like reverse Santas. Robin Hood types." "Robin Hood stole from the rich to give to the poor, not to line his own pockets." "Details, details. Look, are you in or out?" "I don't know, Molly. How would we even pull it off?" "Simple. We dress up as clowns, blend in with the entertainment." "Clowns? At a kid's party? That's so cliché." "Exactly why it'll work. No one will suspect a thing." "And then what? We just grab the presents and run?" "No, dummy. We stuff them in our oversized clown pants, of course." "This is insane. You know that, right?" "Insanely brilliant. Now, can you juggle?" "Juggle? What does that have to do with anything?" "Every good clown needs a act, Bill. Now start practicing."
id
"gen-anvRf8LmDG1Ems7JgpI2LON9Wu5r"
total_cost
0.005235
upstream_id
"msg_01Vup9taCvoGSGeYMfwpT3K7"
created_at
"2024-07-19T15:38:25.99031+00:00"
Friday, July 19th, 2024 at 3:38:25 PM GMT+00:00
model
"anthropic/claude-3.5-sonnet:beta"
app_id
182717
streamed
false
cancelled
false
provider_name
"Anthropic"
latency
12129
moderation_latency
(null)
generation_time
0
finish_reason
"end_turn"
tokens_prompt
68
tokens_completion
290
native_tokens_prompt
75
native_tokens_completion
334
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.005235
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Matches word count
n/a
neededClean
false
words
200
83.6519%
Dialogue to Total Word Ratio
Ratio: 96.50%, Deviation: 6.50%
neededClean
false
wordsTotal
200
wordsDialogue
193
91.8259%