NC Bench
Overview
About
Tests
Models
Model
GPT-4 Turbo
Test
Dialogue tags
Scenario
Write 200 words with 10% dialogue
Run ID
2024-07-19-01J35SM4ACTKYNNMFA816ANB0P
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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
Under the flickering streetlight, Molly sketched out their master plan on a greasy napkin. The scent of pizza wafted through the night air, making the plan seem more like a plot for an after-school special than a famous heist. Bill watched, his face screwed up in concentration or possibly indigestion from the pizza. Molly pointed vigorously at her doodle of a clown and a mountain of presents. "We’ll mix in during the magic show," whispered Molly, careful not to let her words carry beyond the dim light. "You sure those kids won’t notice?" Bill mumbled through a mouthful of crust. "Nah, kids never suspect the funny bunny assistants," Molly assured with a wink. Bill nodded, still unsure but trusting Molly's knack for unconventional plans, where disguises often involved far more glitter than seemed necessary for a heihst. In preparation, they'd practiced balloon animal shaping and even watched three seasons of a children's magic show. Tomorrow, as Molly tied balloon swords and Bill pulled stuffed rabbits from hats, they'd sneak presents out underneath the guise of helping pack up the show. An entire stash of superheroes, dolls, and games awaited them — prime targets for their most whimsical caper yet.
id
"gen-vgHKm7du0vMLDF4wtjyWRymDjcge"
total_cost
0.0083
upstream_id
"chatcmpl-9mjn4oEMTiPfvqDZ4a50XpQHsWHW4"
created_at
"2024-07-19T15:26:41.123554+00:00"
Friday, July 19th, 2024 at 3:26:41 PM GMT+00:00
model
"openai/gpt-4-turbo"
app_id
182717
streamed
false
cancelled
false
provider_name
"OpenAI"
latency
10448
moderation_latency
195
generation_time
0
finish_reason
"stop"
tokens_prompt
68
tokens_completion
254
native_tokens_prompt
68
native_tokens_completion
254
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.0083
Evaluation details
Result
Evaluator
Details
Meta Data
99.9990%
Matches word count
n/a
neededClean
false
words
199
99.9991%
Dialogue to Total Word Ratio
Ratio: 10.55%, Deviation: 0.55%
neededClean
false
wordsTotal
199
wordsDialogue
21
99.9990%