NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1HGNTP4H5EZXNNFQDR43JES
Run Details
Model:
Gemini 3.8 Flash (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
47.1s
Cost:
$0.059349
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 cramped back of the rusty brown van smelled intensely of damp upholstery, stale corn chips, and clumsy amateur espionage. Molly leaned over an intricately crumpled blueprint of house forty two Whispering Pines Lane, tapping a stubby yellow pencil against a crude red circle labeled Danger Zone. Her thick black tactical turtleneck was far too warm for a humid Saturday afternoon, causing beads of nervous sweat to ruin her dramatic makeup. Across from her, Bill sat on an overturned wooden pickle bucket, nervously peeling the foil from a greasy stick of cheap string cheese. "Look at the perimeter." Molly hissed through clenched teeth, her eyes dark with manufactured gravity. "Molly, that is a bouncy castle shaped like a depressed giraffe." Bill squinted through the filthy rear window into the sunlit suburban yard. "It is an elevated observation post. High ground." Bill sighed, chewing a rubbery strand of dairy. "Are we really robbing a little boy, Molly? It feels wrong." Molly slammed her pencil flat against the clipboard with theatrical fury. "Boys with wealthy grandmothers. Do you know what is inside those bags? Cash envelopes. Untraceable joy." Bill considered the undeniable economic logic of fencing pristine building blocks and remote control drones. "What about the mother patrolling the patio?" Molly jabbed a blurry surveillance photograph taped beside the layout. "Brenda. PTA treasurer. Merciless. She guards the artisanal juice station." Bill adjusted his dark sunglasses with a solemn nod. "I can neutralize Brenda. I will ask if the dip contains nightshades. That buys us twenty minutes." Molly smiled thinly, scribbling the diversionary tactic in the margin. "Masterful. While she lectures, I bypass the pinata tripwire." Bill shuddered, gesturing toward a brightly fringed burro swaying from an oak branch. "Do not touch the paper donkey, Molly. It looks entirely unstable." She nodded once, conceding the point with a grim tip of her chin. "Agreed. The real prize sits behind the clown. We extract the loot during the balloon animal workshop." Bill furrowed his brow, wiping cheese grease onto his denim trousers. "Why during the balloon act?" Molly leaned in close, speaking with chilling professional certainty. "Because nobody looks away from a balloon dachshund. The tension is unbearable." Bill stared into the yard, watching the hired entertainer twist purple latex with sweaty, frantic desperation. The poor clown looked completely unhinged beneath his melting greasepaint. The man was right on the brink of snapping, which guaranteed total audience captivity. It was, Bill realized with growing awe and wicked delight, a genuinely flawless window of opportunity. Little Timmy was destined to suffer a tragic loss of remote control racecars, but the suburban underworld offered no sympathy to wealthy toddlers who owned two pony stables. Bill quietly crushed his empty wrapper, slipped a pair of oversized velvet sacks from beneath his crate, and offered an approving nod. "Synchronize your watch, Bill. Operation Sugar Crash commences in precisely three minutes." Molly pulled a black ski mask over her face, exhaled sharply, and clicked open the rusted sliding door.
created_at
"2026-09-02T16:54:14.988Z"
9/2/2026, 4:54:14 PM
model
"google/gemini-3.8-flash-20260902"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1221
moderation_latency
(null)
generation_time
46987
tokens_prompt
75
tokens_completion
4773
native_tokens_prompt
67
native_tokens_completion
15813
native_tokens_completion_images
0
native_tokens_reasoning
15153
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
0.059349
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788368054-xfnzeMI7n7L2Ce73n9WR"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788368054-tWwxDEWloKTJqZqrfOS1"
upstream_id
"t1SYauu_BOG8_uMPhJHN-AI"
provider_responses
0
endpoint_id
"d376f781-516d-42f9-936d-46f19abca6cc"
id
"t1SYauu_BOG8_uMPhJHN-AI"
is_byok
false
latency
1221
model_permaslug
"google/gemini-3.8-flash-20260902"
provider_name
"Google AI Studio"
status
200
total_cost
0.059349
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%