NC Bench
Overview
About
Tests
Models
Model
WizardLM 2 8x22b
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-07-19-01J35SG64TNTTYE8NX0MZ6BEBW
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, focus! We're not here to admire the balloon arches," Bill whispered, adjusting his clown nose. "Sorry, Bill, it's just so... sparkly. Look at that bouncy castle! It's like a fortress of fun." "Forget the castle, we're here for the loot. Now, the presents are by the gift table, right?" "Yep, under that 'Happy Birthday' banner that looks like it's been attacked by a glitter bomb." "Perfect. We'll wait until the kids are cake-faced and distracted, then swoop in." Molly giggled, "Swoop in? You make us sound like superheroes." "Thieves are the unsung heroes of birthday parties, Mol. We liberate toys from the clutches of sticky-fingered tots." "Liberate, huh? So, we're modern-day Robin Hoods in silly hats?" "Exactly. Only instead of giving to the poor, we give to ourselves. Now, I'll create a diversion. You sidle up to the gifts." "What kind of diversion?" "I'll do my world-famous 'Dance of the Sugar Plum Fairy'—on stilts." Molly snorted. "World-famous? You mean infamous." "Semantics. Just be ready to grab and go." "Got it, Bill. But what if we get caught?" "We're not getting caught. We're the dynamic duo of distraction and daring. Besides, who's going to suspect the entertainment?" "You sure about that? Remember the last gig?" "Hey, that wasn't my fault. You can't predict a pie-throwing contest gone wrong." "Okay, okay. Let's do this. But if things go south, I'm blaming your tutu." "Deal. Now, let's make this heist a birthday to remember—for us."
id
"gen-lCmSXRKfJuSDJRJCboDGdWiyVPjF"
total_cost
0.00037548
upstream_id
"cmpl-20a5532c79b641b9ba1dc5ab4f536e39"
created_at
"2024-07-19T15:24:44.204695+00:00"
Friday, July 19th, 2024 at 3:24:44 PM GMT+00:00
model
"microsoft/wizardlm-2-8x22b"
app_id
182717
streamed
false
cancelled
false
provider_name
"DeepInfra"
latency
22475
moderation_latency
(null)
generation_time
0
finish_reason
"stop"
tokens_prompt
95
tokens_completion
381
native_tokens_prompt
112
native_tokens_completion
484
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.00037548
Evaluation details
Result
Evaluator
Details
Meta Data
0.0000%
Matches word count
n/a
neededClean
false
words
251
87.7233%
Dialogue to Total Word Ratio
Ratio: 96.02%, Deviation: 6.02%
neededClean
false
wordsTotal
251
wordsDialogue
241
43.8617%