NC Bench
Overview
About
Tests
Models
Model
Gemini 3 Flash (Preview)
Test
Dialogue tags
Scenario
Write 500 words with 70% dialogue
Run ID
2026-02-08-01KGYX6BR0P2RMSMR0YAGGZDYG
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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, if you suggest the Trojan Horse again, I’m going to shove you into a bouncy castle and leave you there until the heat death of the universe. I’m just saying it’s a classic for a reason, Bill whispered, adjusting his night-vision goggles despite it being three in the afternoon. We get a giant papier-mâché pony. We hide inside. We wait for the cake, then we strike. It’s a six-year-old’s birthday, Bill. The pony will be three feet tall. Unless you’ve mastered the art of molecular compression, you aren’t fitting in a pony. Molly checked her watch, then squinted through the hedges at the suburban fortress before them. A banner reading HAPPY BIRTHDAY KEVIN fluttered mockingly in the breeze. What about the clown move? Bill asked. I put on the shoes, the nose, the whole bit. I make a balloon giraffe, distract the mother, and you bag the loot. You can’t make a balloon giraffe. Last time you tried, you popped three and made a toddler cry so hard he threw up on your boots. That was a technical error with the latex quality, Bill snapped. Focus. The haul is in the sunroom. Twelve Lego sets, three remote-controlled drones, and rumor has it, a limited-edition space station. We go in during the Pin the Tail on the Donkey. The donkey? That’s high-stakes, Molly. Everyone’s looking at the donkey. Exactly. They’re looking at the donkey, not the side door. I’ve got the distraction ready. Bill looked at her suspiciously. Is it the smoke bomb? Please tell me it’s not the smoke bomb. My lungs still hurt from the Bar Mitzvah job. It’s not a smoke bomb. It’s the bees. Bees? Molly, we’re thieves, not supervillains. Where did you even get bees at this hour? They’re mechanical, you idiot. Remote-controlled. They just buzz loudly and hover near the juice boxes. Chaos ensues, the parents panic about anaphylaxis, and we slide in through the laundry room. I don’t like it. It’s too loud. I prefer the silent approach. I could crawl through the vents. The vents are four inches wide, Bill. Are you a hamster? I have high flexibility! You have a hip replacement pending. Now, pull up the blueprints. Bill unfolded a napkin with a crayon drawing of a house. Is this a restaurant napkin? Molly asked. It’s a high-fidelity rendering. Look, here’s the pile of gifts. Here’s the snack table. If we time it right, I can grab a handful of pigs-in-a-blanket on the way out. No snacks! Professionalism, Bill! Do you want the space station or a soggy hot dog? Can’t I have both? It’s a long walk back to the van. Just put on the tactical diaper bag and get the honey-glaze spray ready. The honey-glaze? For the mechanical bees. It makes them more convincing. Molly, I really think the pony was a better idea. One more word about the pony and I’m letting the bees loose in the van. Fine. But if I get tackled by a guy named Gary wearing a 'World’s Best Dad' apron, I’m telling him this was all your idea. Move out, Bill. Kevin’s about to blow the candles. It’s showtime.
created_at
"2026-02-08T15:16:24.525918+00:00"
Sunday, February 8th, 2026 at 3:16:24 PM GMT+00:00
model
"google/gemini-3-flash-preview-20251217"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
814
moderation_latency
(null)
generation_time
7565
tokens_prompt
61
tokens_completion
718
native_tokens_prompt
67
native_tokens_completion
772
native_tokens_completion_images
0
native_tokens_reasoning
0
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
0.0023495
router
(null)
provider_responses
0
status
200
is_byok
false
latency
814
endpoint_id
"6e82f652-2693-4b60-bbe0-7362d6e345f0"
provider_name
"Google AI Studio"
model_permaslug
"google/gemini-3-flash-preview-20251217"
api_type
"completions"
id
"gen-1770563776-qECmf1Nj2P8xvlyCJ1kQ"
upstream_id
(null)
total_cost
0.0023495
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0000%
Matches word count
n/a
neededClean
false
words
541
0.0000%
Dialogue to Total Word Ratio
Ratio: 2.22%, Deviation: 67.78%
neededClean
false
wordsTotal
541
wordsDialogue
12
0.0000%