Model

Claude 3.0 Sonnet

Test

Dialogue tags

Scenario

Write 200 words with 90% dialogue

Run ID

2024-07-19-01J35SG64TTHRF26ENX8716YKQ

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
"Hey, Molly," Bill whispered, leaning against the alleyway wall. "You ready for this gig?" Molly rolled her eyes. "Of course I'm ready. I was born ready for this kind of thing." "Stealing presents from a bunch of kids, though?" Bill scratched his head. "Isn't that a little low, even for us?" "Low?" Molly scoffed. "It's a goldmine, Bill! Those little rugrats get showered with the latest toys and gadgets. We're talking top-of-the-line loot here." "I don't know, Molly. It just feels wrong, you know?" "Oh, come on. It's not like we're stealing from the kids directly. We're just... redistributing their gifts to more appreciative recipients. Like us." Bill shrugged. "If you say so. But if we get caught, I'm blaming you." "Relax, would you? We've got this in the bag. Just follow my lead, and we'll be in and out before anyone even realizes the presents are gone." "Alright, alright. But if we end up on the naughty list, I'm never listening to your crazy schemes again." Molly grinned. "That's the spirit! Now, let's go crash that party and make off with some serious loot!"
id"gen-BplywLh1dHYcOPVLK5x1GQdD0GFq" 
total_cost0.004935
upstream_id"msg_01MWJEi755x9Sc1cx7eddEcN" 
created_at"2024-07-19T15:24:28.537703+00:00" Friday, July 19th, 2024 at 3:24:28 PM GMT+00:00
model"anthropic/claude-3-sonnet:beta" 
app_id182717
streamedfalse
cancelledfalse
provider_name"Anthropic" 
latency6782
moderation_latency(null)
generation_time0
finish_reason"end_turn" 
tokens_prompt68
tokens_completion270
native_tokens_prompt75
native_tokens_completion314
num_media_prompt(null)
num_media_completion(null)
origin"https://ncbench.com/" 
usage0.004935

Evaluation details

Result Evaluator Details Meta Data
81.2727% Matches word count n/a
neededCleanfalse
words188
99.9812% Dialogue to Total Word Ratio
Ratio: 88.83%, Deviation: 1.17%
neededCleanfalse
wordsTotal188
wordsDialogue167
90.6270%