NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KRP49Z5RNSNABJNQFH59DG
Run Details
Model:
Gemini 3.7 Flash (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
32.2s
Cost:
$0.037951
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
100.0000%
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 rain battered the roof of the rusty surveillance van parked two houses down from Elm Street. Inside, Molly adjusted her binoculars, though it was barely two o'clock on a sunny Tuesday afternoon. She tapped a glossy blueprint spread across the dashboard, pinning it flat with a stale jelly donut. "Look at the perimeter. We have heavy defenses. The primary obstacle is an inflatable dragon fortress." Bill squinted over his thermos of tepid instant coffee. He carefully examined the plastic party horn he had modified to look like a dangerous tactical sidearm. "If the dragon has slides, our extraction window drops to ninety seconds. What is the total value of the target payload?" Molly smoothed the paper with professional focus, pointing to a pencil sketch of a decorated plastic picnic table situated directly beneath the wooden patio gazebo. "Liam turns eight today. His wealthy grandmother arrived from Miami, which means expensive robotics and unmarked cash envelopes. The haul should easily clear four figures if we move." Bill stroked his stubbled chin, nodding along with intense, solemn and criminal gravity. "And the hostile presence?" Molly clicked a red ballpoint pen against the cracked passenger window glass. "Twelve sugar crazed children and two parents drinking wine loudly. If a single juice box bursts, this entire operation goes completely hot." They had robbed freight trains in Nevada and jewelers in Manhattan, but neither veteran thief had ever faced a hazard as volatile as a suburban backyard. Bill checked the oversized pockets of his baggy rainbow trousers, ensuring his supply of diversionary confetti was dry, clean, separated, and instantly accessible. "I still think we should have brought smoke bombs instead of bubble soap." Molly glared at him with sharp, lethal exasperation over her cold aluminum clipboard. "We are professionals, Bill. You juggle by the swings to distract the rowdy crowd, while I quickly sweep the gift table and retreat." Bill pulled on his red rubber clown nose, adjusting the tight elastic band carefully. "What about the golden retriever?" Molly patted a pouch on her tactical belt with supreme and truly unshakeable criminal confidence. "Artisan bacon strips. Standard neutralization protocol." Across the street, the front door swung open, revealing an enormous mountain of brightly wrapped presents covered in shiny dinosaur paper. The moment had finally come to execute the grandest suburban heist of their careers. "Synchronize watches. We move on my mark." Bill snapped his oversized polka dot suspenders against his broad chest and grinned with pure, manic criminal joy. "May the party favor us." With practiced precision, the two disguised thieves slipped out the side door of the van, vanishing into the manicured suburban shrubbery like brightly colored ghosts on a secret mission. They crept beneath the wooden fence, fully prepared to ruin the happiest afternoon of eight year old Liam's privileged life, armed only with squeaky shoes, cheap juggling balls, and a ruthless desire for unwrapped toy robots, leaving nothing behind in the abandoned vehicle except empty donut wrappers and sweet, unyielding ambition.
created_at
"2026-09-03T13:52:42.31Z"
9/3/2026, 1:52:42 PM
model
"google/gemini-3.7-flash-20260813"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1460
moderation_latency
(null)
generation_time
32124
tokens_prompt
75
tokens_completion
3328
native_tokens_prompt
67
native_tokens_completion
10107
native_tokens_completion_images
0
native_tokens_reasoning
9478
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
0.0379515
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788443562-6NncVEENc31YxYX2GTzi"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788443562-W0FSc69TPiTEPKRftmx0"
upstream_id
"qnuZat7DG5KSjrEPj6fzmAU"
provider_responses
0
endpoint_id
"dd072012-3274-417d-97e3-e290cba37e31"
id
"qnuZat7DG5KSjrEPj6fzmAU"
is_byok
false
latency
1459
model_permaslug
"google/gemini-3.7-flash-20260813"
provider_name
"Google AI Studio"
status
200
total_cost
0.0379515
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
100.00%
Matches word count
n/a
neededClean
false
words
500
100.00%
Dialogue to Total Word Ratio
Ratio: 30.00%, Deviation: 0.00%
neededClean
false
wordsTotal
500
wordsDialogue
150
100.0000%