NC Bench
Overview
About
Tests
Models
Model
Llama 3.1 8B
Test
Dialogue tags
Scenario
Write 500 words with 50% dialogue
Run ID
2024-07-23-01J3G818BKPZCK50Y52725NET0
Model input
User/Human
You are a author of comedic heist books. Please write a 500 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 planning room was a cramped, smoke-filled back alleyway above a rundown Chinese restaurant. Molly sat hunched over a makeshift table, pouring over a blueprint of the Happy Times children's party venue. Bill leaned against the wall, eyeing the map with a critical gaze. "So, how's the layout?" Bill asked, his voice low and gravelly. Molly looked up, her bright pink hair gleaming in the dim light. "It's a beast. The party hall is one big open space, surrounded by private rooms for the birthday kid and their pals. Total surveillance. We'll need to get in, grab the presents, and get out before anyone raises an alarm." Bill nodded thoughtfully. "I've got a thing for kids' parties. The merchandise is always top-notch. I'm thinking we hit this place during the cake cutting. Everyone's distracted." Molly raised an eyebrow. "You really think they'll let us waltz in with a bunch of screaming five-year-olds running around?" "I didn't say that," Bill said, defensively. "I said we hit during the cake cutting. When the grown-ups are preoccupied." Molly shook her head. "You're going soft, Bill. You used to be a pro." Bill's expression turned wounded. "Hey, I still have it where it counts. My skills are as sharp as ever." Molly snorted. "We'll see about that. Okay, let's get down to business. What's our entry and exit plan?" "I was thinking we go in through the back door," Bill said, pointing to a small window on the blueprint. "I can hotwire the AC unit to get us in. From there, we follow the ventilation shafts to the main hall." Molly scribbled some notes on a piece of paper. "That's 75 feet of crawl space. We're talking serious dust and cobwebs. What's your plan for getting us through that safely?" Bill shrugged. "We'll take turns crawling. I'm sure we can handle a little legwork, Molly." Molly shot him a skeptical look. "A little legwork? You were wheezing like a dying accordion last time we had to crawl through a vent shaft." Bill's face reddened. "That was one time, and it was an emergency! Besides, it's not like we're going up against some highly trained security team here. We're talking nursery school drop-offs." Molly raised a hand, exasperated. "We can't underestimate the powers of rage-filled sugar highs and entitled parent-dom. This is a very specific kind of beast we're dealing with. Don't get cocky." Bill nodded slowly, his expression thoughtful. "Fair point. I'll make sure to keep my legs locked and my wits about me. Now, about the getaway...?" Molly's eyes gleamed with a mischievous spark. "Oh, I've got a sweet ride lined up for us. We'll ride off into the sunset in a minivan, leaving a trail of destruction and disappointed birthday kids in our wake." Bill grinned, rubbing his hands together in anticipation. "I love this plan already."
id
"gen-US9IOXAnK2ktw0GwAmHLTBrwJ1pA"
total_cost
0.0001452
upstream_id
"60044e1e-9cef-4bee-b1f1-df5b15b22c6e"
created_at
"2024-07-23T16:50:47.765048+00:00"
Tuesday, July 23rd, 2024 at 4:50:47 PM GMT+00:00
model
"meta-llama/llama-3.1-8b-instruct"
app_id
182717
streamed
false
cancelled
false
provider_name
"Fireworks"
latency
2624
moderation_latency
(null)
generation_time
0
finish_reason
"stop"
tokens_prompt
102
tokens_completion
654
native_tokens_prompt
70
native_tokens_completion
656
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.0001452
Evaluation details
Result
Evaluator
Details
Meta Data
43.3783%
Matches word count
n/a
neededClean
false
words
483
0.0000%
Dialogue to Total Word Ratio
Ratio: 71.01%, Deviation: 21.01%
neededClean
false
wordsTotal
483
wordsDialogue
343
21.6892%