NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2024-11-04-01JBW8KQKQVK5EGEEM4MT8SJBE
Run Details
Model:
Claude 3.5 Haiku
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
4.6s
Cost:
$0.001172
Evaluation summary
Score
Evaluator
Details Link
43.38%
Matches word count
(details)
93.69%
Dialogue to Total Word Ratio
(details)
68.5317%
Overall
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
The folding table creaked under the weight of Bill's meticulously drawn blueprints of the suburban birthday party venue. Molly leaned in, her red hair falling across her face as she traced a finger along the carefully marked entry points. "These decorations are our perfect cover," Bill muttered, adjusting his glasses. The room was cluttered with surveillance photos of bouncy castles, balloon arches, and tables laden with wrapped gifts. Molly's eyes gleamed with calculated excitement. Children's parties were always chaotic, which meant maximum opportunity and minimal attention. She pointed to a side entrance near the kitchen. "We'll slip in during the piƱata chaos. Parents never watch presents during sugar-induced mayhem." Bill nodded, calculating the potential haul. Superhero-themed wrapping paper suggested expensive electronics and collector's items hidden beneath cartoon characters. Their specialized bags were designed to muffle any accidental rustling. "Forty-five minutes," Bill calculated. "We grab, we go." Molly smirked, already mentally counting their potential profit. The birthday boy would remember this party for entirely different reasons than his parents intended. Just another day in their peculiar profession of opportunistic theft, disguised as social camouflage.
total_cost
0.001172
Evaluation details
Result
Evaluator
Details
Meta Data
43.38%
Matches word count
n/a
neededClean
false
words
183
93.69%
Dialogue to Total Word Ratio
Ratio: 15.05%, Deviation: 5.05%
neededClean
false
wordsTotal
186
wordsDialogue
28
68.5317%