NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KR8Z27EV3XG2CQGQRB76HH
Run Details
Model:
Gemini 3.7 Flash (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
1m 7s
Cost:
$0.026713
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
0.49%
Matches word count
(details)
99.78%
Dialogue to Total Word Ratio
(details)
50.1376%
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 pungent smell of stale pepperoni and damp crushed velvet filled the cramped interior of the rusted surveillance van. Molly tapped the chewed tip of a cherry Twizzler against an elaborate tactical blueprint she had painstakingly sketched on the greasy backside of a Chuck E. Cheese placemat. Across the milk crate table, Bill adjusted a pair of military-grade night-vision goggles, an accessory that was entirely unnecessary and aggressively conspicuous under the blazing glare of a two-o'clock suburban Tuesday afternoon. Bill muttered, "Give me the layout again. I need clarity on hostiles." Molly sighed, blowing a stray curl of dark hair out of her eyes. With theatrical precision, she used a plastic butter knife to point toward a crudely shaded polygon in the northeast corner of the suburban living room, labeled simply The Danger Zone. "Target is Leo’s seventh birthday," she whispered. "The perimeter is crawling with hyperactive sugar-fiends and a discount magician. The loot is staged near the folding chairs." "What kind of haul are we talking?" Bill asked. "Top-shelf merchandise. Three drones, a motorized quad bike, and imported building bricks. The kind parents buy out of guilt." Bill whistled softly, his eyes widening with unvarnished greed behind the thick green glass of his goggles. He reached into a threadbare duffel bag to inventory their specialized infiltration apparatus. Inside lay two conical cardboard party hats, a pair of oversized novelty glasses with rubber noses attached, and an insulated cooler packed with four frozen juice pouches. "Are we properly weaponized?" Bill asked, hefting a pouch. "Blueberry lemonade flavor," Molly confirmed. "Guaranteed to cause a stampede if we need a distraction." She checked her wristwatch, tapping the scratched crystal face. Through the one-way tinted glass of the sliding side door, the target residence looked deceptively peaceful. Yellow balloons bobbed gently along the manicured hedge, while the distant, muffled thump of a rented bouncy castle vibrated through the warm afternoon air like a suburban war drum. "What about the mother?" Bill inquired nervously. "Mrs. Higgins is busy carving an organic ice cream cake. That gives us our window." "And if the magician spots us?" "Tell him his card tricks are derivative. It will shatter his fragile ego." Bill nodded slowly, absorbing the ruthless psychological warfare strategy. He zipped up his dark windbreaker, securing a heavy-duty collapsible canvas laundry hamper across his broad back. They were true professionals, seasoned veterans of chaotic county fairs and cutthroat suburban yard sales, ready to execute the greatest elementary school heist of the modern fiscal year. "Breach in two minutes," Molly declared, sliding the van door open. "No detours for the dessert table." "Not even the funfetti cupcakes?" Bill pleaded. "Eyes on the prize, William. We steal toys, not empty carbohydrates." Molly adjusted her cardboard party hat, checked her elastic chin strap, and led the charge toward the unsuspecting suburban backyard.
created_at
"2026-09-03T13:46:16.708Z"
9/3/2026, 1:46:16 PM
model
"google/gemini-3.7-flash-20260813"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1115
moderation_latency
(null)
generation_time
21067
tokens_prompt
75
tokens_completion
2757
native_tokens_prompt
67
native_tokens_completion
7110
native_tokens_completion_images
0
native_tokens_reasoning
6484
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
0.02671275
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788443176-Gn7vA9mGcpUbTDaIvBBe"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788443176-egVbM4WGwNAtb6tLuTHm"
upstream_id
"KHqZat-LNILUjMcPk8WGuQs"
provider_responses
0
endpoint_id
"dd072012-3274-417d-97e3-e290cba37e31"
id
"KHqZat-LNILUjMcPk8WGuQs"
is_byok
false
latency
1114
model_permaslug
"google/gemini-3.7-flash-20260813"
provider_name
"Google AI Studio"
status
200
total_cost
0.02671275
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
0.49%
Matches word count
n/a
neededClean
false
words
473
99.78%
Dialogue to Total Word Ratio
Ratio: 29.79%, Deviation: 0.21%
neededClean
false
wordsTotal
480
wordsDialogue
143
50.1376%