NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCTJDNMQKRGNS70HGV2BDNH
Run Details
Model:
Gemini 3.6 Flash (Reasoning, Minimal)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
6.0s
Cost:
$0.005328
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 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 for seven-year-old Timmy’s birthday party was drawn in melted orange crayon on a soggy IHOP placemat. Molly smoothed it out, her dark eyes gleaming with the manic intensity of a seasoned mastermind. "Step one," Molly whispered, tapping a spot labeled Bouncy Castle. "We secure the perimeter." Bill squinted through his battered binoculars at the suburban chaos across the street. "Molly, there’s a mother wearing a visor and wielding a glitter gun. She looks unhinged." "Focus, Bill! This is the motherload. The snitch in pre-K confirmed Timmy’s getting the Lego Death Star, three Nerf blasters, and a heavily padded envelope from his wealthy grandmother." Bill sighed, adjusting his black turtleneck in the ninety-degree summer heat. "I still think stealing from a child is bad karma. Plus, I’m allergic to latex. If a balloon pops near my face, I’m going into anaphylactic shock." "Where is your professional pride?" Molly snapped, pulling two oversized party hats out of her duffel bag. "We aren't stealing from a child. We're liberating high-end plastic from a pint-sized bourgeois dictator who doesn't even appreciate fine engineering. Put your cone on." "It squashes my ears," Bill complained, dutifully tying the silver elastic band under his double chin. "What’s our cover identity?" "We’re the sub-contracted entertainment." "Wait, did you book us as clowns?" "Worse. We’re ambient balloon artists." Bill gasped, nearly dropping his binoculars. "I don't know how to make a dog!" "You twist the sausage shape until it squeaks," she said coldly. "Now, here’s the play. At fourteen hundred hours, the juice box distribution creates a tactical bottleneck near the patio. Parents will be distracted by sugar-induced hyperventilation. That’s when you strike the gift table." "And what’s your role?" "I’m taking out the pinata." "The pinata isn't target cargo, Molly!" "It’s filled with king-sized candy bars, Bill! I’m not leaving chocolate on the table. We swept a bar mitzvah last month and you let six pounds of fudge get impounded by the caterers. Never again." Bill checked his glittery cartoon watch. "Two minutes to drop time. Should we double-check the escape route?" "The slip-n-slide is greased and ready," Molly confirmed, adjusting her dark sunglasses. "If things go sideways, we breach the azalea bushes and rendezvous at the ice cream truck." "If I get sticky, I’m quitting the guild," Bill muttered, pulling a pair of tiny safety scissors from his tactical belt. Molly grabbed him by his ruffled collar, her face inches from his. "Look at me. This is for the ultimate prize. Forty-eight boxes of unboxing joy. Do you want that Death Star or not?" "I want to build the superlaser," Bill admitted softly. "Then move out." They darted across the lawn, keeping low behind the flamingo lawn ornaments, ready to commit the fluffiest crime of the century.
created_at
"2026-07-25T14:23:51.484Z"
7/25/2026, 2:23:51 PM
model
"google/gemini-3.6-flash-20260721"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
540
moderation_latency
(null)
generation_time
5971
tokens_prompt
75
tokens_completion
749
native_tokens_prompt
67
native_tokens_completion
697
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
0.005328
router
(null)
provider_responses
0
endpoint_id
"5d6d133d-a953-4781-a835-7ee76f4e1388"
id
"98ZkaqjpI7aN_uMPx5qi2A8"
is_byok
false
latency
540
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-1784989431-0BKWxB4N3ZFfPaaoXbnz"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784989431-CJE0iDe8atsqeRL6hdOn"
upstream_id
"98ZkaqjpI7aN_uMPx5qi2A8"
total_cost
0.005328
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
462
0.00%
Dialogue to Total Word Ratio
Ratio: 65.54%, Deviation: 35.54%
neededClean
false
wordsTotal
473
wordsDialogue
310
0.0000%