NC Bench
Overview
About
Tests
Models
Model
Gemini 3 Pro (Preview)
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2025-11-19-01KACPTZDWWKQ2E4MN2V512FKW
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
The blueprint, Molly said, unrolling a grease-stained napkin. It’s a bouncy castle, Mol. A fortress of solitude and vinyl. We hit them during the Happy Birthday song. Audio cover is essential. That’s thirty seconds, tops. I can’t run that fast. Not in these clown shoes. You have to. The payload is sitting right next to the cake. A mountain of toy trucks. And the guard dog? Mr. Fluffles is easily bribed. Did you bring the jerky? I ate it. I got nervous. You ate the distraction? I’m a stress eater! Besides, the real threat is the magician. He sawed a lady in half, Molly. In half. Smoke and mirrors, Bill. Focus. I repel down the drainpipe, snag the gifts, and toss them to you in the bushes. What if a toddler intercepts? Deploy the countermeasures. The bubbles? The bubbles. Aim for the eyes. It confuses their depth perception. It seems ruthless. It’s a heist, Bill. Not a playdate. Is the getaway vehicle primed? My mom’s minivan is idling around the corner. The sliding door sticks, though. We’ll make it work. Put on the mask. I can’t breathe in this SpongeBob thing. Suffocate later. We have loot to liberate.
created_at
"2025-11-19T00:04:58.0181+00:00"
Wednesday, November 19th, 2025 at 12:04:58 AM GMT+00:00
model
"google/gemini-3-pro-preview-20251117"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2038
moderation_latency
(null)
generation_time
16659
tokens_prompt
68
tokens_completion
965
native_tokens_prompt
67
native_tokens_completion
1864
native_tokens_completion_images
0
native_tokens_reasoning
1567
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
0.022502
api_type
"completions"
id
"gen-1763510681-UThcnjoGSyQU14PPDc8Q"
upstream_id
(null)
total_cost
0.022502
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
Evaluation details
Result
Evaluator
Details
Meta Data
99.9990%
Matches word count
n/a
neededClean
false
words
199
0.0000%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 90.00%
neededClean
false
wordsTotal
199
wordsDialogue
0
49.9995%