NC Bench
Overview
About
Tests
Models
Model
Llama 3.1 Nemotron 70B
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-10-16-01JABJ0P4JSSG1HV005EGFGN0K
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
"I'm telling you, Bill, this is the score of a lifetime," Molly said, unfolding a crude hand-drawn map of "Bouncy Castle World". "A children's birthday party? You want to steal from kids?" Bill raised an eyebrow, his voice laced with skepticism. "Hey, those kids are gonna get more presents next year. We're just... redistributing wealth," Molly justified, a sly grin spreading across her face. "Redistributing into our pockets, you mean. What's the haul look like?" "According to my sources – aka the heavily caffeinated birthday mom's Facebook posts – we're talking at least twenty iPad Minis, a handful of American Girl dolls, and a guaranteed slew of cash-filled cards." Bill whistled low. "Not bad. But what's the plan for getting past the bounce house guards?" "Bribery, my friend. I've got a stash of premium cupcakes with extra sprinkles. No exhausted parent can resist." "And the, you know, actual kids? Won't they, I don't know, notice us stealing all their stuff?" Molly waved her hand dismissively. "Pfft, they'll be too busy screaming on the ball pit to even notice."
id
"gen-1729112332-ZUdTinmKCEyAZOkp2eNB"
upstream_id
"cmpl-RM3vgGURZbZAX2y69cy5Y1fP"
total_cost
0.00012485
cache_discount
(null)
created_at
"2024-10-16T20:59:01.374242+00:00"
Wednesday, October 16th, 2024 at 8:59:01 PM GMT+00:00
model
"nvidia/llama-3.1-nemotron-70b-instruct"
app_id
182717
streamed
true
cancelled
false
provider_name
"DeepInfra"
latency
106
moderation_latency
(null)
generation_time
8143
finish_reason
"stop"
tokens_prompt
102
tokens_completion
251
native_tokens_prompt
71
native_tokens_completion
250
native_tokens_reasoning
(null)
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.00012485
Evaluation details
Result
Evaluator
Details
Meta Data
9.6081%
Matches word count
n/a
neededClean
false
words
178
41.8184%
Dialogue to Total Word Ratio
Ratio: 80.34%, Deviation: 9.66%
neededClean
false
wordsTotal
178
wordsDialogue
143
25.7133%