NC Bench
Overview
About
Tests
Models
Model
GPT-4o, Aug. 6th (temp=1)
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-08-06-01J4MXY7FJV1A4JCBKYPJEH35H
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
"Alright, Bill. What's the plan for the party heist?" "Molly, it's a simple snatch and grab. No fuss." "But it's a children's party, Bill. We need to blend in." "Blend in? Easy. We dress as clowns." "Clowns? Oh, come on! I can't juggle, Bill!" "No need to juggle. Just smile and act ridiculous." "Ridiculous I can do. But how do we handle all the screaming kids?" "We bring balloons. Lots of balloons." "Balloons as distraction? Clever, Bill." "Exactly, Molly. While you make balloon animals, I swipe the presents." "And what about the parents? They have eyes, you know." "We'll improvise. Maybe throw a pie." "Throw a pie? Seriously?" "Fine, no pie. We’ll offer them cupcakes." "Those mini ones that nobody really wants?" "Precisely. They’ll be too busy deciphering sprinkles." "Okay, then? We’re in, grab the loot, and out?" "Yes, quick and easy. In and out." "What if we get caught?" "We won't. We're the best, Molly." "Okay, Bill. But if this goes south, we join the circus." "Deal. Now let's go get those presents." "Alright, let's try not to steal any kids along the way." "Agreed, Molly. Let’s keep this heist strictly gift-related."
id
"gen-nxXZ43WHPGAHaDd4UahiKd7zuEca"
total_cost
0.0031275
upstream_id
"chatcmpl-9tNEUIPZBypak7plJxgS3wdYoifLw"
created_at
"2024-08-06T22:46:17.647241+00:00"
Tuesday, August 6th, 2024 at 10:46:17 PM GMT+00:00
model
"openai/gpt-4o-2024-08-06"
app_id
182717
streamed
false
cancelled
false
provider_name
"OpenAI"
latency
3478
moderation_latency
172
generation_time
0
finish_reason
"stop"
tokens_prompt
68
tokens_completion
307
native_tokens_prompt
67
native_tokens_completion
296
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.0031275
Evaluation details
Result
Evaluator
Details
Meta Data
98.7124%
Matches word count
n/a
neededClean
false
words
194
36.7879%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
194
wordsDialogue
194
67.7502%