NC Bench
Overview
About
Tests
Models
Model
GPT-4.1
Test
Dialogue tags
Scenario
Write 200 words with 50% dialogue
Run ID
2025-04-14-01JRTSZYE2V9P5XKSEYYRD2XQQ
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 50% 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
Molly crouched behind the inflatable unicorn, peering through a gap in the bouncy castle’s pastel flank. “Are you sure this is the right party? Last time you almost stole someone’s gluten-free cupcakes by mistake.” Bill smirked, clutching a glittery duffel bag. “That’s because you said, ‘follow the rainbow balloons,’ and apparently, every kid’s party has rainbow balloons.” Molly rolled her eyes. “Look, it says ‘Happy Seventh Birthday, Tyler!’ on that banner. We want Tyler. He’s loaded. His aunt’s a toy executive.” “Perfect,” Bill whispered. “So, what’s the plan? Ninja up the slide or blend in with the face-painters?” “I tried face painting last time. The kids kept asking for dinosaurs and I only know how to do abstract clowns.” “That explains the nightmares, then,” Bill shuddered. “No, this time, we infiltrate the magician's act. Everyone watches the hat—no one watches the loot.” Molly grinned. “So, you distract with some questionable magic and I sweep the gift table?” “Exactly. Just one thing—don’t accidentally saw me in half again.” “No promises if you keep talking.” Bill straightened his striped vest. “Alright. Let’s go be the least mature adults here.” “You mean, second least mature. Tyler’s uncle is still dancing.”
id
"gen-1744656399-0rSMOo7pLoANoyZbdJ16"
upstream_id
"chatcmpl-BMJ7H0r3A6qVqgM9xlX5qDmltYUHE"
total_cost
0.002454
cache_discount
(null)
provider_name
"OpenAI"
created_at
"2025-04-14T18:46:43.919539+00:00"
Monday, April 14th, 2025 at 6:46:43 PM GMT+00:00
model
"openai/gpt-4.1-2025-04-14"
app_id
182717
streamed
true
cancelled
false
latency
373
moderation_latency
124
generation_time
3347
tokens_prompt
68
tokens_completion
289
native_tokens_prompt
67
native_tokens_completion
290
native_tokens_reasoning
0
num_media_prompt
(null)
num_media_completion
(null)
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.002454
Evaluation details
Result
Evaluator
Details
Meta Data
99.9990%
Matches word count
n/a
neededClean
false
words
201
0.0000%
Dialogue to Total Word Ratio
Ratio: 81.09%, Deviation: 31.09%
neededClean
false
wordsTotal
201
wordsDialogue
163
49.9995%