NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KS1DASDY72TWWXK72HV2EK
Run Details
Model:
Gemini 3.7 Flash (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
13.5s
Cost:
$0.014642
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
91.24%
Dialogue to Total Word Ratio
(details)
45.6210%
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 cramped surveillance van smelled intensely of stale cheese crackers and theatrical villainy. Molly smoothed out a blueprint sketched in neon crayon across the back of a greasy paper placemat. She wore full black tactical gear, complete with heavy night-vision goggles that were entirely useless in the blinding two o'clock afternoon glare. Bill sat on an overturned cooler, casually chewing a strawberry licorice twist while cleaning his fingernails with the tip of a plastic spoon. Molly slammed a laser pointer against the blueprint, illuminating a wobbly drawing of an inflatable bounce house. "Look at the perimeter," Molly whispered fiercely. "The target is the sunroom." "Why are we whispering?" Bill asked. "We are fifty yards away." "Professionalism requires stealth," Molly hissed, tapping the drawing. "Intelligence says there are forty high-end parcels inside." Outside the tinted windows, the chaotic shrieks of hyperactive children echoed down the quiet suburban driveway. A towering mountain of gift boxes wrapped in metallic foil shimmered through the bay window like gold bars in a bank vault. Molly adjusted her oversized tactical vest, which was currently stuffed with zip-ties, distraction juice boxes, and several miniature smoke pellets. "What about the guard?" Bill asked, nodding toward the garden. "Mr. Chuckles," Molly replied darkly. "Do not look into his eyes. He is fueled entirely by cold brew and balloon animal rage." "I hate clowns," Bill said. "What if the mother sees us?" "Karen will be distracted," Molly said with absolute certainty. "She is currently struggling to divide an organic cake evenly among twenty toddlers. Her cognitive load is fully compromised." Bill squinted at the crayon schematic, rubbing his stubbled chin with obvious skepticism. He had participated in high-stakes museum robberies and European diamond heists, but the sheer unpredictability of thirty sugar-crazed eight-year-olds unnerved him far more than laser tripwires or armed bank guards ever could. "And the getaway?" Bill asked. "Through the doggie door," Molly said. "Then over the wooden fence." "I had four burritos for lunch," Bill pointed out, frowning. "I will get stuck." "You will butter your hips if necessary," Molly stated without a hint of humor. "Those video games are ours." Bill sighed, reached into his tactical pouch, and quietly loaded a neon plastic slingshot with sticky rubber frogs. The stakes had never been lower, yet Molly’s terrifying intensity made it feel like they were about to breach an impenetrable underground military bunker. "Should I deploy the glitter bombs?" Bill asked. "Only as a last resort," Molly said grimly. "Move out."
created_at
"2026-09-03T13:58:51.999Z"
9/3/2026, 1:58:51 PM
model
"google/gemini-3.7-flash-20260813"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1321
moderation_latency
(null)
generation_time
13420
tokens_prompt
75
tokens_completion
1449
native_tokens_prompt
67
native_tokens_completion
3891
native_tokens_completion_images
0
native_tokens_reasoning
3329
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
0.0146415
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788443932-QZIGuDQndvPq4jVMpvrd"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788443932-qu8n5b9SfGmZtsiSkhEL"
upstream_id
"HH2ZaomyCfLUjMcPt_veuQY"
provider_responses
0
endpoint_id
"dd072012-3274-417d-97e3-e290cba37e31"
id
"HH2ZaomyCfLUjMcPt_veuQY"
is_byok
false
latency
1321
model_permaslug
"google/gemini-3.7-flash-20260813"
provider_name
"Google AI Studio"
status
200
total_cost
0.0146415
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
0.00%
Matches word count
n/a
neededClean
false
words
414
91.24%
Dialogue to Total Word Ratio
Ratio: 31.35%, Deviation: 1.35%
neededClean
false
wordsTotal
421
wordsDialogue
132
45.6210%