Model

GPT-4.1

Test

Dialogue tags

Scenario

Write 200 words with 90% dialogue

Run ID

2025-04-14-01JRTSZG6FQTB4RX62H3D0VEZ5

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

AI
“Molly, you got the map?” “Right here, Bill. See this? The unicorn piñata? That’s our entry.” “You sure you want to climb inside a piñata?” “Do you want to dodge ten six-year-olds dressed as Spider-Man?” “Fair. So after the unicorn?” “We crawl out behind the cake table, grab the goodie bags, then the presents, then out the back window.” “Wait. The clown is stationed by the window.” “I’m not afraid of clowns.” “That’s a lie and you know it.” “Fine. We use the doggy door.” “You got stuck last time.” “I was wearing a tutu! I’m in black this time.” “Alright, so we ninja-roll through the doggy door. What if the dog’s home?” “We feed it the leftover meatballs from the buffet.” “Smart. Now, the presents. How do we not get trampled?” “Distraction. You pretend you’re the magician.” “I don’t know any magic.” “You just need to shout ‘Watch closely!’ and make things vanish.” “Like what?” “Like the pile of presents, if you’re fast enough.” “You believe in me too much, Molly.” “I believe in expensive wrapping paper. Let’s go steal some.” “You bring the rope?” “Bill, always.”
id"gen-1744656384-i6NUQQNdFX8C9cBv0N7r" 
upstream_id"chatcmpl-BMJ72pofcEkkaMZf1lJGwvohdod1L" 
total_cost0.002502
cache_discount(null)
provider_name"OpenAI" 
created_at"2025-04-14T18:46:28.73259+00:00" Monday, April 14th, 2025 at 6:46:28 PM GMT+00:00
model"openai/gpt-4.1-2025-04-14" 
app_id182717
streamedtrue
cancelledfalse
latency453
moderation_latency98
generation_time2816
tokens_prompt68
tokens_completion295
native_tokens_prompt67
native_tokens_completion296
native_tokens_reasoning0
num_media_prompt(null)
num_media_completion(null)
num_search_results(null)
origin"https://ncbench.com/" 
is_byokfalse
finish_reason"stop" 
native_finish_reason"stop" 
usage0.002502

Evaluation details

Result Evaluator Details Meta Data
95.9868% Matches word count n/a
neededCleanfalse
words192
36.7879% Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededCleanfalse
wordsTotal192
wordsDialogue192
66.3873%