NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2025-11-19-01KACPMN7HC75GJP045WHWH0YG
Run Details
Model:
Gemini 3 Pro (Preview)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
34.6s
Cost:
$0.055550
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
99.38%
Matches word count
(details)
99.67%
Dialogue to Total Word Ratio
(details)
99.5218%
Overall
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
Bill unrolled the stolen blueprints across the diner booth’s sticky formica surface with dramatic flair. It was actually a crudely rendered crayon drawing of the Peterson’s backyard, snatched from their refrigerator earlier. He looked intense, treating the wrinkled paper like the complex schematics for a high-security bank vault. "Target acquisition is fourteen hundred hours," Bill whispered, strictly adjusting his sunglasses. Molly sighed, slowly stirring her lukewarm black coffee to delay looking at him. "The clown arrives at two, Bill. Just say two." "Precision is key. The perimeter is guarded by a Golden Retriever named Buster. He has a weakness for peanut butter." Molly patted her heavy coat pocket. "I bought the chunky kind." Bill nodded solemnly, tracing a gloved finger over a red wax squiggle. "Excellent. The payload is on the picnic table. We extract the gift bags during the piñata frenzy." Molly stared at the markings, rubbing her aching temples. "You realize you are planning a tactical Black Ops insertion into a five-year-old's birthday party?" "These aren't just kids, Molly. They are tiny, sticky surveillance drones. If we get made, we face the ultimate interrogation." "Tantrums," Bill shuddered. "We grab the loot and vanish." "If we get caught?" Bill smirked. "We blame the magician."
created_at
"2025-11-19T00:01:49.312345+00:00"
11/19/2025, 12:01:49 AM
model
"google/gemini-3-pro-preview-20251117"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2457
moderation_latency
(null)
generation_time
34470
tokens_prompt
68
tokens_completion
1780
native_tokens_prompt
67
native_tokens_completion
4618
native_tokens_completion_images
0
native_tokens_reasoning
4326
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
0.05555
api_type
"completions"
id
"gen-1763510473-wNAryIJ02GQc33gBcYXf"
upstream_id
(null)
total_cost
0.05555
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
Evaluation details
Result
Evaluator
Details
Meta Data
99.38%
Matches word count
n/a
neededClean
false
words
205
99.67%
Dialogue to Total Word Ratio
Ratio: 52.40%, Deviation: 2.40%
neededClean
false
wordsTotal
208
wordsDialogue
109
99.5218%