NC Bench
Overview
About
Tests
Models
Model
Gemini 3 Pro (Preview)
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2025-11-19-01KACR4JQSFHMK6SZW2WYDS61D
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
Check the perimeter again, Molly hissed. It’s a backyard, Molly. The perimeter is a white picket fence. It’s a containment zone, Bill. Don't get complacent. What is the intel on the target? A mountain of brightly wrapped boxes guarding the chocolate fountain. The primary objective appears to be a limited edition starship set. Excellent. High resale, low volume. And the security detail? A dozen six-year-olds currently sugar-crashing and one drowsy dad manning the propane grill. Deadly. Those spatulas can be weaponized. We need a distraction. I still think we should just pull the fire alarm. Too loud. We stick to the plan. I release the feral hamster into the bouncy castle. In the ensuing chaos, you navigate the crepe paper streamers and secure the package. I have serious questions about the extraction vehicle. The red wagon is strategically parked behind the hydrangeas. It has a squeaky wheel, Molly. It adds authenticity. We act like helpful neighbors moving gifts to the car. We blend in. We are wearing balaclavas. They are themed! Mine has ears. I’m a bandit mouse. I look like a generic burglar. You look like a party favor gone wrong. Stop whining. Do you want the starship or not? I’m just saying, if a toddler bites me, I’m telling on him. Just put on the party hat. It’s go time.
created_at
"2025-11-19T00:27:47.4591+00:00"
Wednesday, November 19th, 2025 at 12:27:47 AM GMT+00:00
model
"google/gemini-3-pro-preview-20251117"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1988
moderation_latency
(null)
generation_time
18137
tokens_prompt
68
tokens_completion
997
native_tokens_prompt
67
native_tokens_completion
1933
native_tokens_completion_images
0
native_tokens_reasoning
1615
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
0.02333
api_type
"completions"
id
"gen-1763512048-EnMiThdUQnk21ZBp4794"
upstream_id
(null)
total_cost
0.02333
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
Evaluation details
Result
Evaluator
Details
Meta Data
2.0116%
Matches word count
n/a
neededClean
false
words
225
0.0000%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 90.00%
neededClean
false
wordsTotal
225
wordsDialogue
0
1.0058%