NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1HJD3D7907A5WBA4K4GN0XM
Run Details
Model:
Gemini 3.8 Flash (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
19.8s
Cost:
$0.024703
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
97.63%
Matches word count
(details)
94.86%
Dialogue to Total Word Ratio
(details)
96.2420%
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 interior of the rusted surveillance van smelled intensely of stale onion rings, lukewarm gas-station coffee, and high-stakes suburban ambition. Molly pinned a blurry telephoto printout to the corkboard with decisive aggression. In the exact center of the frame was the primary target: a towering mountain of brightly wrapped boxes piled high on a redwood patio table. Beside it, a seven-foot inflatable lawn dragon swayed violently in the erratic autumn breeze, menacing the perimeter with plastic fangs. Molly stood before the layout like a seasoned general, tapping her red grease pencil against the cork. Across the narrow aisle, Bill squinted through a pair of smudged binoculars, carefully adjusting the focus wheel with a trembling thumb. He surveyed the manicured battlefield with mounting dread. A massive bouncy castle hissed in the distance, pulsing like an ominous, neon-colored heart. "Look at the perimeter security, Molly. We are talking two golden retrievers and an uncle who clearly wrestled in college." Molly did not even look up from her clipboard. She was busy drawing precise approach vectors across a satellite map of the cul-de-sac. "Those dogs are currently unconscious on a pile of discarded hot dog buns. Uncle Steve has already downed three rum punches. His reaction time is zero." "What about the sugar-crazed children?" "The puppet show begins at two sharp. It is a sock-based drama about dental hygiene that will render every human under the age of forty catatonic for twenty-two minutes." Bill lowered the binoculars and rubbed his temples in despair. He stared out at the pastel balloons tied to the white mailbox. On the corrugated steel floor between them lay three empty industrial laundry sacks. He had spent fifteen respectable years cracking commercial bank safes and boosting luxury sedans, but here he was, hiding in a flower-delivery van and plotting against an elementary schooler. The sheer absurdity of the gig was threatening to break his spirit entirely. "Timmy is turning eight today. Are we genuinely stooping to boosting remote-control drones and high-end plastic building blocks?" "The aftermarket markup on collector sets is astronomical, Bill. That plastic is untraceable currency." She reached into a frayed duffel bag, rummaged past several rolls of duct tape, and produced a pair of bulbous, scarlet clown shoes, dropping them heavily onto his lap. "Why on earth am I wearing squeakers?" "If someone spots you in the rhododendrons, you are not a felon. You are simply the entertainment. It is called tactical camouflage." "They honk with every single step." "Then walk lightly, you amateur." Bill sighed deeply and wrestled the oversized rubber footwear over his steel-toed boots. The rubber toes were comically rounded, flapping uselessly against the metal pedals. A pathetic, rubbery squeak echoed through the cramped cabin. Through the tinted rear window, the cheerful, tinny melody of an accordion began drifting over the manicured hedges. The show was starting. Molly pulled down her dark aviator sunglasses, checked her stopwatch with theatrical precision, and clicked the van doors open.
created_at
"2026-09-02T17:24:26.158Z"
9/2/2026, 5:24:26 PM
model
"google/gemini-3.8-flash-20260902"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2413
moderation_latency
(null)
generation_time
19781
tokens_prompt
75
tokens_completion
2478
native_tokens_prompt
67
native_tokens_completion
6574
native_tokens_completion_images
0
native_tokens_reasoning
5929
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
0.02470275
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788369866-ZCgf6KbQpJ79LrfCfDx0"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788369866-TLzX4xgxhuWlNKcQP9Jn"
upstream_id
"yluYar7IDcugjMcP77ep-QY"
provider_responses
0
endpoint_id
"d376f781-516d-42f9-936d-46f19abca6cc"
id
"yluYar7IDcugjMcP77ep-QY"
is_byok
false
latency
2413
model_permaslug
"google/gemini-3.8-flash-20260902"
provider_name
"Google AI Studio"
status
200
total_cost
0.02470275
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
97.63%
Matches word count
n/a
neededClean
false
words
493
94.86%
Dialogue to Total Word Ratio
Ratio: 31.03%, Deviation: 1.03%
neededClean
false
wordsTotal
506
wordsDialogue
157
96.2420%