NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCT14A573FVPYFGQZVWBW1A
Run Details
Model:
Gemini 3.6 Flash (Reasoning, Minimal)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
5.9s
Cost:
$0.005088
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 50% 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 adjusted her night-vision goggles, which were currently resting on a pair of oversized, glittery fake eyelashes. "I still say we go through the bouncy castle," Bill whispered, unrolling a detailed blueprint of a suburban backyard onto the hood of their dented hatchback. "It’s the path of least resistance." "Bill, it’s inflatable. One snag from your tactical pocketknife and we’re jumping on a deflating, squeaking mattress in front of twenty sugary eight-year-olds," Molly said, tapping her finger on the blueprint. "We go through the patio doors. Clean, quiet, and away from the clown." "The clown is off-duty by two o'clock," Bill argued, crossing his arms. "I checked his schedule. He has a kid's birthday in Glendale at three." "Clowns are unpredictable, Bill. You know this. It's Heist Rule Number Four." "Rule Four is never trust a man with two first names!" "It applies to clowns too!" Molly hissed, ducking as a neighborhood cat strolled past the car. "Now, focus. What is the payload?" "The loot is stacked on the kitchen island," Bill said, pulling a tiny notebook from his vest. "According to my intel, we are looking at three LEGO Star Wars sets, a motorized scooter, and a box wrapped in shiny dinosaur paper that feels suspiciously like a Nintendo Switch." "Dinosaur paper?" Molly’s eyes widened. "That’s high-value targets only. What about the decoy?" "I bought a bag of frozen broccoli and a pair of tube socks," Bill replied proudly. "If we swap them out, the birthday boy won't even notice until we're halfway to the border." "The border? Bill, we live in Ohio." "It's the principle of the thing, Molly. We need a dramatic exit strategy." "Fine," Molly sighed, checking her watch. "It’s one-forty-five. The piñata massacre should be starting any minute now. That’s our distraction." "Are you sure about the sugar-crash timing?" "Positive. They’ve consumed three gallons of blue punch and an entire ice cream cake. When that piñata bursts, it’s going to be absolute lord-of-the-flies chaos. Nobody will be watching the gift table." "Except the mother," Bill noted grimly. "Carol. She’s a hawk." "I have a countermeasure for Carol." Molly reached into her duffel bag and pulled out a thermos. "Single-origin pumpkin spice latte. Extra foam." Bill gasped, looking at her with newfound respect. "You didn't." "Oh, I did. She won't be able to resist. She'll be incapacitated by seasonal cozy vibes for at least ten minutes." "You're a evil genius," Bill whispered. "I'm a professional," Molly corrected, pulling down her ski mask. "Now grab the burlap sacks. It’s time to ruin a seven-year-old's Tuesday."
created_at
"2026-07-25T14:14:25.338Z"
7/25/2026, 2:14:25 PM
model
"google/gemini-3.6-flash-20260721"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
494
moderation_latency
(null)
generation_time
5326
tokens_prompt
75
tokens_completion
682
native_tokens_prompt
67
native_tokens_completion
665
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
0.005088
router
(null)
provider_responses
0
endpoint_id
"5d6d133d-a953-4781-a835-7ee76f4e1388"
id
"wcRkasGaGM_h_uMPpoT0mQc"
is_byok
false
latency
494
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-1784988865-QnFEmo7sJuWPJPbvoHkG"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784988865-Max8lWVBTabQWiGHGeVg"
upstream_id
"wcRkasGaGM_h_uMPpoT0mQc"
total_cost
0.005088
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
428
0.00%
Dialogue to Total Word Ratio
Ratio: 75.11%, Deviation: 25.11%
neededClean
false
wordsTotal
442
wordsDialogue
332
0.0000%