NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KRVZQN699KF1YYC08XKDQV
Run Details
Model:
Gemini 3.7 Flash (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
27.1s
Cost:
$0.029465
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
86.38%
Matches word count
(details)
99.90%
Dialogue to Total Word Ratio
(details)
93.1414%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 stale scent of warm vinyl upholstery and artificial lemon air freshener filled the cramped interior of the rusted surveillance van. Molly tapped the laminated blueprint of the suburban backyard with the chewed end of an unsharpened crayon, her expression as grim as a seasoned general facing imminent battlefield catastrophe. Outside, a neon-green bouncy castle swayed violently beneath the chaotic weight of twelve over-sugared kindergarteners. It was zero hour on Maple Street, and the suburban defenses were crumbling under the relentless siege of festive anarchy. Molly adjusted her binoculars and whispered, "Pay attention, Bill. The drop zone is located between the juice box cooler and the patio slider." Beside her, Bill shifted uncomfortably on an overturned plastic milk crate. He was dressed in a stifling, full-body purple unicorn costume complete with sparkly hooves and an enormous plush mane, breathing heavily through the tiny mesh mouth slit while wiping condensation from the plastic interior. "Remind me why I am wearing this unicorn outfit? The horn keeps hitting the rearview mirror." "Because children respect mythical quadrupeds," Molly replied, without looking away from the patio. "If a toddler spots you hauling an electric go-kart across the lawn, they will assume you are an enchanted courier." She focused the high-powered lenses on the screened back porch. The prize table was a dazzling masterpiece of domestic excess and conspicuous consumption. Towering pyramids of shiny metallic foil wrapping paper and giant satin bows concealed what their neighborhood informant confirmed to be several high-end robotic toys, a top-tier gaming console, and an imported motorized scooter. Timmy Johnson was turning seven years old today, and his wealthy hedge-fund godparents had clearly spared no expense in attempting to purchase his fleeting affection. Bill wiped a bead of sweat from his synthetic muzzle. "What about the mother? She looks like she benches two hundred pounds." "Mrs. Johnson is trapped at the glitter station. Her hands are covered in craft glue. She poses zero tactical threat." "And the clown?" "Professor Giggles is a lightweight," Molly muttered dismissively. "Two squeaks of his nose and he flees to the punch bowl." "I hate toddlers," Bill grumbled, checking the heavy nylon sack hooked to his sequined belt. "They have sticky fingers and no concept of personal space." "Stick to the plan," Molly commanded, checking her tactical wristwatch. "When the candy hits the grass, the children will riot. That is our cue to breach the fence, bag the loot, and vanish." A sharp, splintering crack echoed across the manicured yard as a blindfolded six-year-old wielding a wooden bat shattered the paper donkey into colorful fragments. Screaming children descended upon the spilled cascade of hard candies like a pack of rabid piranhas fighting over fresh meat. The suburban perimeter was entirely broken. Molly racked the plastic slide on a heavy-duty party-popper, zipped her utility vest, and gave Bill the definitive nod. The ultimate, most ridiculous suburban heist of their disgraced criminal careers had officially begun.
created_at
"2026-09-03T13:55:55.668Z"
9/3/2026, 1:55:55 PM
model
"google/gemini-3.7-flash-20260813"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1584
moderation_latency
(null)
generation_time
25553
tokens_prompt
75
tokens_completion
3244
native_tokens_prompt
67
native_tokens_completion
7844
native_tokens_completion_images
0
native_tokens_reasoning
7212
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"STOP"
service_tier
"default"
usage
0.02946525
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788443755-F4V9G7hDqeGsKfNkoZEw"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788443755-54S06yBu95GoRd0cKIMB"
upstream_id
"a3yZatezMdTV1MkPsfLnoA8"
provider_responses
0
endpoint_id
"dd072012-3274-417d-97e3-e290cba37e31"
id
"a3yZatezMdTV1MkPsfLnoA8"
is_byok
false
latency
1584
model_permaslug
"google/gemini-3.7-flash-20260813"
provider_name
"Google AI Studio"
status
200
total_cost
0.02946525
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
86.38%
Matches word count
n/a
neededClean
false
words
489
99.90%
Dialogue to Total Word Ratio
Ratio: 30.14%, Deviation: 0.14%
neededClean
false
wordsTotal
501
wordsDialogue
151
93.1414%