NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCWV2XCJDK8JKB3KJCRCCYM
Run Details
Model:
Gemini 3.6 Flash (Reasoning, Minimal)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
5.7s
Cost:
$0.004803
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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 gestured frantically at the laminated floor plan spread across the hood of their beat-up sedan. Look at the perimeter, Bill. It is a minefield. We are robbing an eight-year-old, Molly, not Fort Knox, Bill said, rubbing his temples. Why are we treating a bouncy castle like a SAM missile site? Because that bouncy castle blocks our primary exfiltration route, she snapped, pointing at a cartoon dragon drawing. If Timmy’s sugar-crazed friends initiate a bounce-pocalypse, our getaway is blown. Fine, what is the entry strategy then? We disguise ourselves as the entertainment, Molly whispered, leaning in close. I managed to acquire authentic uniforms from a contact in the prop industry. Please tell me we are not clowns, Bill groaned. I do not do red noses. My face breaks out. Worse, Molly said grimally. Face painting artists. We carry the toxic glitter and the washable markers. That is diabolical, Bill said, a slow smile spreading across his face. Nobody suspects the face painter. It is the ultimate Trojan horse. Precisely, she said. Once inside the perimeter, we identify the high-value targets. We are looking for large boxes, metallic wrapping paper, and anything that heavy-clinks like high-end Lego sets. What about the hostiles? Bill asked, tapping the diagram. The matriarch looks formidable. Karen, Molly confirmed. She has operated three juice box stations simultaneously. Her situational awareness is off the charts. You will distract her with a complex question about gluten-free catering while I move the loot. And the clown? Bill pointed at a scribbled circle labeled Bozo. Is he armed? Only with balloon animals and squeaky shoes, Molly said. But he is unpredictable. If he spots us double-bagging the Xbox, we drop a jar of organic apple juice to create a slip hazard and make for the fence. I still think we should just snatch the pile during the pinata, Bill muttered, crossing his arms. It is classic chaos theory. Everyone looks up at the paper donkey, we look down at the Nerf blasters. Too risky, Molly countered. Pinata blindfolds create erratic melee damage. You could take a stray broomstick to the kneecap. We stick to the tactical glitter operation. What is the timetable? Cake is served at fourteen hundred hours, Molly said, checking her watch. The sugar rush peaks at fourteen fifteen. That gives us a seven-minute window of absolute anarchy before the inevitable meltdowns begin. Seven minutes to clear out forty presents, Bill calculated. We need big bags. I stole trash bags from a municipal dump, she replied proudly. Heavy duty. Scented like vanilla to blend in with the baking smells. You are a criminal mastermind, Molly. I know, she said, pulling two rainbow aprons from the backseat. Now put on your glitter holster. It is time to make some money.
created_at
"2026-07-25T15:03:32.533Z"
7/25/2026, 3:03:32 PM
model
"google/gemini-3.6-flash-20260721"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
528
moderation_latency
(null)
generation_time
5584
tokens_prompt
75
tokens_completion
716
native_tokens_prompt
67
native_tokens_completion
627
native_tokens_completion_images
0
native_tokens_reasoning
0
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
0.004803
router
(null)
provider_responses
0
endpoint_id
"5d6d133d-a953-4781-a835-7ee76f4e1388"
id
"RNBkarGdKNL1jrEPrPjGgAI"
is_byok
false
latency
527
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-1784991812-lltTgCPLfkTcFKXaoIvr"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784991812-U3X1ZdqtIOeMnjwX0HWY"
upstream_id
"RNBkarGdKNL1jrEPrPjGgAI"
total_cost
0.004803
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
0.00%
Matches word count
n/a
neededClean
false
words
461
0.00%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 70.00%
neededClean
false
wordsTotal
472
wordsDialogue
0
0.0000%