NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCSD4SSGBWXC8AP51TQNV4M
Run Details
Model:
Gemini 3.6 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
32.8s
Cost:
$0.060656
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
93.65%
Matches word count
(details)
99.69%
Dialogue to Total Word Ratio
(details)
96.6682%
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 blueprint laid out on the ping-pong table was drawn entirely in washable crayon, yet Molly analyzed it with the terrifying intensity of a veteran art thief. She adjusted her night-vision goggles, which were currently perched awkwardly over her glittery forehead, and clicked a red laser pointer against a wobbly circle labeled Zone Red. The basement hummed with the sound of a laundry dryer, adding an eerie, mechanical background noise to their top-secret tactical briefing. "Target acquired," Molly whispered, her voice gravelly with fake tactical grit. "Timmy turns six at fourteen-hundred hours. We have twenty minutes between the pinata and cake cutting to extract the package." Bill sighed loudly, leaning against a stack of stolen folding chairs in the dim corner. He rubbed his aching temples with both thumb and forefinger, deeply questioning every single life choice that had brought him to this damp cellar on a Tuesday afternoon. Outside, the suburban neighborhood was quiet, completely unaware of the criminal mastermind operating right next door. "Molly, it is a bounce house in suburban Ohio, not Fort Knox. We are talking about LEGOs and a plush alpaca." Molly slammed her gloved fist onto the green table, making a loose juice box jump two inches into the air. She leaned forward, her eyes narrowing behind her plastic tactical visor. "It is a high-value haul, Bill! Do you know what a mint Star Wars ship gets on the black market? The kid is six. He does not appreciate fine plastic engineering." Bill crossed his arms, staring at the elaborate poster board. Molly had meticulously mapped out guard positions, patrol routes, and environmental hazards. The danger zones were marked in bright purple and included Sticky Fingers, Unpredictable Tantrums, and a terrifyingly detailed drawing of a Golden Retriever named Barnaby. "What about Mrs. Higgins?" Bill asked, gesturing to the perimeter. "She tackled a mailman last week for walking on hydrangeas." Molly smiled, a cold and calculating expression that looked absurd under her festive surroundings. She reached into her utility belt and dramatically pulled out a shiny, foil-wrapped sphere. "I prepared a diversion," she declared proudly. "A decoy pinata stuffed with organic raisins. While the small heathens suffer disappointment, we breach the patio." Bill winced, taking a slow step backward. "That is an act of war. Those children will riot." "Chaos is our ally." Molly unrolled a heavy-duty, velvet-lined duffel bag across the ping-pong table and zipped open the main compartment with a sharp flourish. "We load the loot, bypass the bouncy castle, and fence the Nerf blasters by nightfall. Are you in?" Bill looked at the velvet bag, then at Molly, whose ridiculous enthusiasm was ultimately impossible to resist. He reached into his leather coat and pulled out a pair of plush clip-on cat ears. "I want first dibs on the Transformers." "Deal. Operation Birthday Blitz is a go." Molly snapped her tactical party hat securely into place beneath her chin, adjusted her elbow pads, and gave a sharp nod. It was finally time to execute the greatest birthday party robbery in suburban history.
created_at
"2026-07-25T14:03:30.216Z"
7/25/2026, 2:03:30 PM
model
"google/gemini-3.6-flash-20260721"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1224
moderation_latency
(null)
generation_time
32425
tokens_prompt
75
tokens_completion
2845
native_tokens_prompt
67
native_tokens_completion
8074
native_tokens_completion_images
0
native_tokens_reasoning
7405
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
0.0606555
router
(null)
provider_responses
0
endpoint_id
"5d6d133d-a953-4781-a835-7ee76f4e1388"
id
"MsJkasmWFZ-5sOIP6ejsgAo"
is_byok
false
latency
1224
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-1784988210-0GUMqo7afGO9MbYnInJ7"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784988210-mhbWmTkcohUmoVVYUOpc"
upstream_id
"MsJkasmWFZ-5sOIP6ejsgAo"
total_cost
0.0606555
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
93.65%
Matches word count
n/a
neededClean
false
words
509
99.69%
Dialogue to Total Word Ratio
Ratio: 30.25%, Deviation: 0.25%
neededClean
false
wordsTotal
519
wordsDialogue
157
96.6682%