NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCYKK25GVHBEFRGQ5B149TF
Run Details
Model:
Gemini 3.6 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
32.5s
Cost:
$0.058390
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
0.01%
Matches word count
(details)
72.40%
Dialogue to Total Word Ratio
(details)
36.2027%
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
Molly slammed a hand-drawn blueprint onto the kitchen table, scattering half-eaten graham crackers across the sticky laminate surface. The dim light of their cramped basement hideout flickered erratically, casting long, dramatic shadows over stolen tactical gear, two discarded capes, and an embarrassing amount of stolen sugar packets. Bill leaned over the map, squinting through the gloom at the vibrant red crayon lines depicting a sprawling two-story suburban house. A massive black X was marked directly over the living room gift table, right next to a greasy thumb smudge that smelled suspiciously like artificial apple juice. On the left side of the paper, Molly had meticulously rendered a bouncy castle complete with tiny, panicked stick figures trapped inside. "This is it, Bill," Molly whispered, her eyes wild with unhinged ambition. "The target is seven years old today. His name is Timmy, and his parents are rich." Bill adjusted his dark, oversized sunglasses, despite being indoors in a windowless room. He sighed, adjusting the straps of a tactical harness that was currently holding three plastic squirt guns and a pair of neon-pink safety goggles. "Are we really doing this, Mol? Stealing from a kid? That feels like a new low, even for us." Molly slapped her hand onto the table, making the rogue graham cracker crumbs jump into the air. She adjusted her tactical headset, which was currently plugged into absolutely nothing, and narrowed her eyes with absolute intensity. "Focus," she snapped, tapping a dull yellow pencil against the crayon map. "Timmy is receiving a brand-new game console, a high-tech drone, and four unboxed Lego sets. The bounty is immense, but security is tight. We are dealing with twelve hyperactive kids, a sugar rush, and a stressed mother armed with juice boxes." Bill scratched his chin, studying the scribbled perimeter. He tapped a drawn fence line with a calloused finger, genuinely impressed by the sheer insanity of her intelligence gathering. "What is our entry point?" he asked. Molly leaned in close, her breath smelling of cheap espresso. "The back patio," she said. "At fourteen hundred hours, a clown named Sprinkles arrives to distract the perimeter. While he fails at balloon animals, we breach the sliding glass door." Bill reached deep into his tactical vest, rummaging past several half-eaten granola bars and a squeaky dog toy before pulling out two bright blue plastic cylinders with pulled pins. "I still say we use smoke bombs," he muttered. Molly folded her arms, raising an eyebrow in suspicion. "Do you have them?" she asked. Bill held up the blue plastic cylinders. "Glitter bombs, actually. Smoke was completely out of stock." A slow, terrifying grin spread across Molly’s face as she snatched one of the canisters, weighing it expertly in her hand. "Even better," she smirked. "Nothing strikes terror into suburban parents quite like craft glitter." Outside their basement window, the distant, muffled sounds of a party horn squealed through the sunny suburban neighborhood air. It was time to strike. They grabbed their oversized burlap sacks, pulled down their ridiculously thick knitted tactical balaclavas, and slipped out into the afternoon heat, absolutely determined to execute the grandest, most absurd gift heist the local parent-teacher association had ever seen.
created_at
"2026-07-25T15:34:24.079Z"
7/25/2026, 3:34:24 PM
model
"google/gemini-3.6-flash-20260721"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1319
moderation_latency
(null)
generation_time
32379
tokens_prompt
75
tokens_completion
3473
native_tokens_prompt
67
native_tokens_completion
7772
native_tokens_completion_images
0
native_tokens_reasoning
7071
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
0.0583905
router
(null)
provider_responses
0
endpoint_id
"5d6d133d-a953-4781-a835-7ee76f4e1388"
id
"gNdkarjGCPLTjMcP5Zr64As"
is_byok
false
latency
1317
model_permaslug
"google/gemini-3.6-flash-20260721"
provider_name
"Google AI Studio"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.18.0; linux; x64))"
http_referer
(null)
request_id
"req-1784993664-REtdw0qmsGwGxKkgO1Cw"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784993664-3lULNIGAsQXlXvQNAnYg"
upstream_id
"gNdkarjGCPLTjMcP5Zr64As"
total_cost
0.0583905
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
0.01%
Matches word count
n/a
neededClean
false
words
531
72.40%
Dialogue to Total Word Ratio
Ratio: 27.46%, Deviation: 2.54%
neededClean
false
wordsTotal
539
wordsDialogue
148
36.2027%