Model

GPT-4 Turbo

Test

Dialogue tags

Scenario

Write 200 words with 90% dialogue

Run ID

2024-07-19-01J35SJ1D4MXMFAK1B3WVK5DWK

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 peeked over the fence, clipboard in hand. "Okay, Bill, plan’s simple. We crash the party, blend in, snatch the goodies." Bill, disguised in a too-tight superhero costume, scoffed. "Blend in? Look at me! I'm a six-foot-two Power Ranger." "Relax, it's a costume party. Now, when the cake comes out, everyone will be distracted. That's our moment." "Right, so I'll grab the presents while you... what? Help serve cake?" "Exactly," Molly grinned, adjusting her clown wig. "Who’d suspect a clown?" Bill rolled his eyes. "Great, the muscle and the circus reject. Flawless." "Not just any clown, Bill. A clown with a plan. We’ll pile the gifts by the back gate." "And getaway?" "The old Johnson van. Discreet, roomy, and it even smells like dirty laundry." "Perfect for a couple of thieving party crashers," Bill muttered. "But what if someone spots us?" "That’s where improvisation comes in. You juggle." "I can’t juggle." "Then just throw things around. It’s chaos, Bill. Live a little!" "Fine, but if a kid cries, I’m out. I’ve got standards." Molly chuckled. "Let’s make this the best party they’ll never remember."
id"gen-sBjGofdF7m6MxtmlJy9mWsobR0Hn" 
total_cost0.00926
upstream_id"chatcmpl-9mjlyNX8Jmc7uoKGHs03dFSKLLZ0B" 
created_at"2024-07-19T15:25:32.392577+00:00" Friday, July 19th, 2024 at 3:25:32 PM GMT+00:00
model"openai/gpt-4-turbo" 
app_id182717
streamedfalse
cancelledfalse
provider_name"OpenAI" 
latency10193
moderation_latency242
generation_time0
finish_reason"stop" 
tokens_prompt68
tokens_completion286
native_tokens_prompt68
native_tokens_completion286
num_media_prompt(null)
num_media_completion(null)
origin"https://ncbench.com/" 
usage0.00926

Evaluation details

Result Evaluator Details Meta Data
75.1556% Matches word count n/a
neededCleanfalse
words187
82.9296% Dialogue to Total Word Ratio
Ratio: 83.42%, Deviation: 6.58%
neededCleanfalse
wordsTotal187
wordsDialogue156
79.0426%