NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCSDWVSA6EEB3D0GBJ6EC1J
Run Details
Model:
Gemini 3.6 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
20.5s
Cost:
$0.035643
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
1.48%
Dialogue to Total Word Ratio
(details)
50.7372%
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 unrolled the blue crayon map of Timmy’s seventh birthday bash across the hood of their dented hatchback. "Look at this perimeter," Bill whispered, pointing a trembling finger. "Juice boxes to the north, inflatable bouncy castle to the south. It is an impenetrable suburban fortress." "Get a grip, Bill," Molly said, adjusting her night-vision goggles in broad daylight. "We are seasoned master thieves, not amateur clown victims. The primary objective is in the living room." "The gift table," Bill gasped, his eyes widening. "A mountain of shiny foil paper, curly silk ribbons, and wrapped Lego sets. It’s breathtaking." "Precisely. At fourteen hundred hours sharp, our tactical distraction begins." "The pinata?" "The pinata," Molly confirmed. "While seventeen sugar-crazed second graders beat a papier-mache donkey with wooden bats, we slip silently through the rear sliding glass door." "What about the mother?" Bill asked, shivering at the thought. "Mrs. Gable is completely ruthless. I saw her hand out sugar-free gluten-free cupcakes last year without even flinching." "I’ve neutralized the threat," Molly replied calmly. "I posted an anonymous tip on the neighborhood Facebook group about a flash sale on artisanal organic Kombucha three blocks away. She will be lured away for at least twenty minutes." "Brilliant," Bill said, nodding in genuine awe. "Absolute tactical genius. But what about the hyper-aggressive guard dog?" "Mr. Fluffington?" Molly smirked. "The three-pound Pomeranian who lacks upper teeth? I brought a slice of premium smoked gouda. He will be eating out of my hand within five seconds." "Okay, fair enough," Bill said, pacing beside the car. "So we breach the perimeter, secure the high-value loot, and bag the wrapped packages. But how do we carry forty presents out to the street without looking suspicious to the neighbors?" "We wear the deep-cover infiltration disguise," Molly said. Bill froze and stared at her. "Tell me it’s not the mascot suits." "It’s definitely the mascot suits." "Molly, please, no. I can’t fit into Barnaby the Stuffed Bear again. The hidden zipper digs directly into my left kidney." "Do you want that battery-powered shiny red fire truck with the actual working siren or not, Bill?" "I really do," Bill admitted quietly. "The box says ages four and up, and I am thirty-four, so I qualify." "Then put on the oversized plush head," Molly commanded, tossing him a massive brown fur mask. "You are Barnaby the Bear. I am Penelope the Polka-Dot Pony. We make our grand entrance during the cake cutting ceremony. While the parents sing, we load the heavy burlap sacks." "What if a small child asks Barnaby for a hug?" Bill asked nervously. "You give them a firm corporate handshake and inform them that Barnaby is strictly unionized," Molly said. "No physical contact without hazard pay." "Understood," Bill said, pulling the bear head down. His voice muffled instantly. "Is it wrong that I feel totally exhilarated?" "It’s the pure thrill of high-stakes heist work, Bill," Molly said, slipping on her pony mask. "Now grab the sacks. It’s time to rob a child."
created_at
"2026-07-25T14:03:54.626Z"
7/25/2026, 2:03:54 PM
model
"google/gemini-3.6-flash-20260721"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1285
moderation_latency
(null)
generation_time
20451
tokens_prompt
75
tokens_completion
2184
native_tokens_prompt
67
native_tokens_completion
4739
native_tokens_completion_images
0
native_tokens_reasoning
4010
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
0.035643
router
(null)
provider_responses
0
endpoint_id
"5d6d133d-a953-4781-a835-7ee76f4e1388"
id
"SsJkasaQKvXe_uMPhbvXwQo"
is_byok
false
latency
1285
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-1784988234-X2Bg5EsItP9Eo3KWPjRL"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784988234-qXrLQeM1FyCqDFuLZQza"
upstream_id
"SsJkasaQKvXe_uMPhbvXwQo"
total_cost
0.035643
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
100.00%
Matches word count
n/a
neededClean
false
words
501
1.48%
Dialogue to Total Word Ratio
Ratio: 79.18%, Deviation: 9.18%
neededClean
false
wordsTotal
514
wordsDialogue
407
50.7372%