NC Bench
Overview
About
Tests
Models
Model
Gemini 2.5 Pro
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2025-11-18-01KACP9G3NQ165P40A3RERBJ9Q
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
Messages
Meta
AI
“Okay, Bill. Remind me of the infiltration plan.” “Right. The clown arrives at three. I knock him out, I take the costume.” “Gently. You knock him out gently. We need the wig intact.” “Right. Gently. I enter the house as… what was his name?” “Jiggles, Bill. You are Jiggles, the delightful party entertainer.” “Jiggles, the Bringer of Birthday Doom.” Molly sighed. “No. Just Jiggles. You distract the children with…?” “The 'ten-legged spider of chaos' balloon animal.” “Excellent. While they're mesmerized, I slip in the back and grab the haul.” “The haul being a remote-controlled car and several LEGO sets.” “It's high-value plastic, Bill. The best kind. What about the piñata?” “I snag it on the way out. You think it's filled with candy or untraceable bearer bonds?” “It's a unicorn, Bill. It's candy.” “A man can dream. And the primary security system?” “She's named Princess Fluffybutt. A miniature poodle. She's our exit.” “We throw her at the dad?” “No. You give her the squeaky hedgehog. It buys us precisely twelve seconds.” “Seems risky. What if she’s a squeaky squirrel kind of dog?” “Then you improvise.”
created_at
"2025-11-18T23:55:23.688729+00:00"
Tuesday, November 18th, 2025 at 11:55:23 PM GMT+00:00
model
"google/gemini-2.5-pro"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1924
moderation_latency
(null)
generation_time
14758
tokens_prompt
68
tokens_completion
826
native_tokens_prompt
67
native_tokens_completion
1750
native_tokens_completion_images
0
native_tokens_reasoning
1442
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"STOP"
usage
0.01758375
api_type
"completions"
id
"gen-1763510108-UA8NwCdAIfuvz35Gttve"
upstream_id
(null)
total_cost
0.01758375
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
Evaluation details
Result
Evaluator
Details
Meta Data
81.2727%
Matches word count
n/a
neededClean
false
words
188
52.8515%
Dialogue to Total Word Ratio
Ratio: 98.94%, Deviation: 8.94%
neededClean
false
wordsTotal
188
wordsDialogue
186
67.0621%