NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCSEA22E0G4A5MV35CHA1G0
Run Details
Model:
Gemini 3.6 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
45.8s
Cost:
$0.091180
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
35.00%
Matches word count
(details)
93.65%
Dialogue to Total Word Ratio
(details)
64.3284%
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 spread the blueprint across the grease-stained coffee table with the solemnity of a general planning a major military invasion. Bill leaned over the map, squinting through a cloud of stale powdered sugar from a half-eaten jelly donut. "Is that a bouncy castle?" Bill asked, pointing a sticky finger at a crude yellow square drawn near the inflatable swimming pool. "That is Ground Zero," Molly corrected, tapping a sharp red mark beside a hand-drawn sandbox. "The target is Timmy Vance. Age seven. Spoiled rotten, highly uncooperative, and guarded by hyperactive toddlers." Bill muttered, scratching his stubbled chin with a look of deep, genuine concern. "I don't know, Molly. Stealing from a kid feels low, even for us." Molly slammed her fist onto the cluttered table, shaking the cold coffee cups violently. "The flamingos had it coming, Bill! Focus. Timmy’s parents bought him three thousand dollars in unopened LEGO sets, a drone, and a pony." "They got him a whole pony?" "A mini pony. It fits in the back of our van if we fold down the rear seats." She pulled a pair of heavy night-vision goggles from beneath the worn leather sofa and tossed them into a black duffel bag alongside two glittery paper cone hats and a cardboard jug. "What's the juice for? Poison?" Bill asked, picking up the heavy container and sniffing it cautiously. "It's organic apple juice," Molly snapped, forcefully jamming a tiny pink glittery birthday hat over his ears. "We need to blend in, idiot. You look like a fun-loving uncle named Barnaby who lost his shoes in the ball pit. Now, listen up. At fourteen hundred hours, the hired clown initiates balloon animals. That's our window." "What if the clown spots us?" "Clowns are colorblind, Bill. It's a scientific fact. Once he distorts reality with rubber swords, we strike." Bill adjusted the tight elastic band cutting into his double chin and sighed deeply, staring down at the suburban layout. "How do we handle the mother? She looked terrifying on the recon footage." "Mrs. Vance will be incapacitated by sugar-free mimosas. I bribed the caterer with a stolen air fryer." "Brilliant, Molly. And the loot?" "We load the wrapped boxes onto the red wagon. If anyone asks, we are transporting the goods to the magical gift vault." "Does the mini pony go in the wagon too?" "The pony gets a carrot and rides shotgun." Bill rubbed his hands together, his early moral objections completely erased by the tactical genius of the operation. He reached inside his leather coat and proudly brandished a shiny silver plastic object. "I brought our heavy tactical backup," he whispered slyly, wiggling his thick bushy eyebrows back and forth. "Is that a bubble gun?" Molly stared at him in complete and utter disbelief, slapping her forehead with a heavy groan. "It creates a defensive smokescreen!" "It creates slippery hazards, Bill!" "Exactly! Tactical slipping!" "Fine, but if you drop soap on my boots, I'm leaving you behind." Molly zipped the heavy duffel bag shut, strapped on her cardboard party mask, and grinned like an ambitious criminal mastermind ready for birthday cake.
created_at
"2026-07-25T14:04:08.14Z"
7/25/2026, 2:04:08 PM
model
"google/gemini-3.6-flash-20260721"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1672
moderation_latency
(null)
generation_time
45661
tokens_prompt
75
tokens_completion
5885
native_tokens_prompt
67
native_tokens_completion
12144
native_tokens_completion_images
0
native_tokens_reasoning
11415
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
0.0911805
router
(null)
provider_responses
0
endpoint_id
"5d6d133d-a953-4781-a835-7ee76f4e1388"
id
"WMJkaorNFKuX-8YPndmxsAk"
is_byok
false
latency
1672
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-1784988248-WSE6vQkduYoO2ObtxqBv"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784988248-yscx3gGp7rGEkwR4u7jh"
upstream_id
"WMJkaorNFKuX-8YPndmxsAk"
total_cost
0.0911805
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
35.00%
Matches word count
n/a
neededClean
false
words
518
93.65%
Dialogue to Total Word Ratio
Ratio: 51.15%, Deviation: 1.15%
neededClean
false
wordsTotal
524
wordsDialogue
268
64.3284%