NC Bench
Overview
About
Tests
Models
Model
Llama 3.1 8B
Test
Dialogue tags
Scenario
Write 500 words with 70% dialogue
Run ID
2024-07-23-01J3G80JBWB28VZ04NNA6NCNVV
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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 leaned back in her chair, eyeing Bill with a mixture of skepticism and amusement. "You can't be serious, Bill. A children's party?" Bill shrugged, his expression deadpan. "Hey, it's a goldmine. Kids' parties are like little wrapped-up treasure chests, just begging to be plundered." Molly raised an eyebrow. "And what exactly do you plan to do with all those presents? Sell them on the black market to tweeters?" Bill waved his hand dismissively. "Details, details. We'll figure it out. The point is, it's a low-risk, high-reward gig. And I've got a guy who's willing to provide us with the uniforms." Molly snorted. "You mean the ones that say 'Crazy Clown' across the front? Yeah, that's going to be a real deterrent for security." Bill grinned. "Hey, it's a start. And think about it, Molly, what kid is going to suspect a couple of clowns coming into a party full of screaming kids and balloons?" Molly sighed, rubbing her temples. "You're not taking this seriously enough, Bill. What if we get caught? What if there's a bouncy castle?" Bill chuckled. "Oh, come on, Molly. You're not still worried about the bouncy castle are you? It's a prop, not a piece of deadly machinery." Molly shot him a dark look. "You don't understand, Bill. Children's parties are like war zones. There are pinatas, and balloons, and possibly hallucinogenic cotton candy...it's like a nightmare come to life." Bill held up his hands, laughing. "Okay, okay, I get it. You're scared of the cotton candy. But what about the presents? We could make a killing. I'm talking kidney-busting, life-changing cash." Molly hesitated, her expression softening ever so slightly. "Okay, fine. But if we get caught, I'm blaming you. And we're not going in till the cake is cut. I don't want to risk our lives for a bunch of half-eaten cupcakes." Bill grinned, pumping his fist in the air. "That's the spirit, Molly! Now let's get planning. I'll get the clown costumes and you can start casing the joint...or rather, the party." Molly stood up, smoothing out her jacket. "And what exactly do I need to do?" Bill scribbled down a list on a pad of paper. "I need you to hit the party planner website, find out what kind of decorations they have, and whether or not there's going to be a DJ." Molly raised an eyebrow, taking the list from Bill. "And why do you need all this information?" Bill winked. "Because, my friend, you need to get into that party planner's head. Know what's going to distract them, what kind of security they'll have, and what kind of cake they'll be serving. That's how we get away with the goods, Molly." Molly rolled her eyes good-naturedly, shaking her head. "You're impossible, Bill. But hey, at least you're consistent." Bill chuckled, collecting the list from Molly. "Hey, that's what they pay me for."
id
"gen-9IJAbLRTAVkOpLbH4YCdJ4voKtL7"
total_cost
0.0001518
upstream_id
"23c8abbc-1e78-49c1-af10-05425d7f8031"
created_at
"2024-07-23T16:50:25.252428+00:00"
Tuesday, July 23rd, 2024 at 4:50:25 PM GMT+00:00
model
"meta-llama/llama-3.1-8b-instruct"
app_id
182717
streamed
false
cancelled
false
provider_name
"Fireworks"
latency
2674
moderation_latency
(null)
generation_time
0
finish_reason
"stop"
tokens_prompt
102
tokens_completion
687
native_tokens_prompt
70
native_tokens_completion
689
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.0001518
Evaluation details
Result
Evaluator
Details
Meta Data
86.3803%
Matches word count
n/a
neededClean
false
words
489
8.0641%
Dialogue to Total Word Ratio
Ratio: 77.10%, Deviation: 7.10%
neededClean
false
wordsTotal
489
wordsDialogue
377
47.2222%