NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KRYG4JFKFJNYM9TYD3YHME
Run Details
Model:
Gemini 3.7 Flash (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
15.4s
Cost:
$0.020687
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
99.9999%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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 bulky night-vision goggles over floral camouflage face paint while Bill unrolled a grease-stained blueprint of the Johnson backyard. The afternoon sun beat down mercilessly on their cramped surveillance van, which smelled perpetually of stolen string cheese and nervous sweat. Outside, thirty sugar-crazed children formed a chaotic perimeter around a towering mountain of brightly wrapped birthday boxes. Molly tapped the plastic inflatable castle. "The bounce house is the blind spot." Bill nodded gravely, chewing stale black licorice as he carefully checked the carrying capacity of their velvet sacks. He had spent three agonizing days tracking the party itinerary, cross-referencing juice-box distribution schedules with the amateur magician’s routine. Security was tight, managed entirely by a suspicious teenager wearing a filthy yellow duck costume. "Watch out for the clown. He's vicious." Molly loaded a customized launcher with soft, noise-dampening marshmallows. This was the ultimate heist of the suburban season, boasting high-end drones, robotic dogs, and three rare gaming consoles. She cracked her knuckles, completely ready to infiltrate the noisy yard before the giant unicorn pinata shattered into colorful pieces. "Bag the Legos first." "Copy that." Bill racked the slide on his tactical confetti cannon, hoisted the empty sack, and charged.
created_at
"2026-09-03T13:57:16.57Z"
9/3/2026, 1:57:16 PM
model
"google/gemini-3.7-flash-20260813"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1371
moderation_latency
(null)
generation_time
15304
tokens_prompt
75
tokens_completion
1612
native_tokens_prompt
67
native_tokens_completion
5503
native_tokens_completion_images
0
native_tokens_reasoning
5234
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
0.0206865
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788443836-ELuDt4TMg7WJXWwXWZMr"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788443836-CSUaBs7ReqMLvEKHM5pm"
upstream_id
"vHyZap6GK7PdjMcP-9rN2A8"
provider_responses
0
endpoint_id
"dd072012-3274-417d-97e3-e290cba37e31"
id
"vHyZap6GK7PdjMcP-9rN2A8"
is_byok
false
latency
1371
model_permaslug
"google/gemini-3.7-flash-20260813"
provider_name
"Google AI Studio"
status
200
total_cost
0.0206865
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
200
100.00%
Dialogue to Total Word Ratio
Ratio: 9.66%, Deviation: 0.34%
neededClean
false
wordsTotal
207
wordsDialogue
20
99.9999%