NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KRQ501AEZCP7Y6EDBBW057
Run Details
Model:
Gemini 3.7 Flash (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
11.6s
Cost:
$0.011195
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
43.38%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
21.6892%
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 spread the crumpled paper across the dashboard and tapped the page with a manicured finger. "Look at the blueprints, Bill." "Molly, that is a drawing of a pony done in purple glitter crayon." "It is a tactical schematic. The pony represents the bouncy castle. If we breach the northwest perimeter at fourteen hundred hours, the guards will be utterly compromised." "Perimeter guards?" "The mothers, Bill. Wine-drunk suburban mothers in high-end yoga pants. They have a massive blind spot near the hibiscus bushes while discussing pilates." Bill adjusted his binoculars and squinted through the tinted windshield. "And the primary target?" "Timmy Jenkins is turning seven. His grandfather owns a hedge fund. We are talking tier-one birthday loot. Four remote-controlled drones, a custom electric go-kart, and eighteen boxes of limited-edition space Legos." "What about the chocolate fountain?" "Leave the fountain, Bill. We are professional thieves, not plumbers. The high-yield boxed merchandise is stacked on the mahogany gift table directly behind the patio slider." "What about the hired muscle?" "Boffo the Clown." "He looks unhinged, Molly. Did you see the size of his shoes? You could conceal a tactical crowbar in those." "Boffo takes a smoke break behind the garage precisely at fourteen-fifteen. That gives us an eight-minute window before the cake cutting ceremony." "Eight minutes? To clear fifty wrapped packages?" "Fifty-four packages, according to the verified RSVP list on Mrs. Jenkins's public Pinterest board." Bill nervously wiped his palms on his overalls. "We still need a diversion." Molly unzipped her duffel bag with a sharp yank. "I brought a burlap sack of live ferrets." "Ferrets? Inside a crowded bouncy castle?" "Children love wildlife, Bill. It creates instantaneous pandemonium. While the panic spreads, we slip past the slider disguised as balloon artists." "I do not know how to twist a balloon animal, Molly. I can only make a vaguely menacing sausage." "Then make menacing sausages and tell them it is abstract expressionism. What matters is the wheelbarrow." "The reinforced wheelbarrow is already staged behind the compost bin. But what if little Timmy spots us?" "Timmy will be in a profound sugar coma after three slices of funfetti cake. If he stirs, we inform him we are Santa's repo division." "In the middle of July?" "Mid-year audits happen, Bill. The kid is seven, not a tax attorney." "What about the home defense system?" "A golden retriever named Barnaby. Highly corruptible. I loaded three pounds of artisanal cocktail sausages into my tactical vest." "Good. I despise dogs with morals. What is our exfiltration route?" "Through the side gate, straight down the service alley to the getaway van." "And if Boffo pursues us?" "Deploy the pocket glitter." "Pocket glitter? That is practically a war crime, Molly." "This is the big leagues, Bill. Those Legos are retiring us to Boca." "Let us review. You enter as the balloon specialist. I breach via the compost with the wheelbarrow. We neutralize the dog, ignore the melted chocolate, bag the electronics, and vanish before the piƱata drops." Molly handed him a red foam ball. "Precisely. Put on your clown nose. We have a birthday to ruin."
created_at
"2026-09-03T13:53:15.783Z"
9/3/2026, 1:53:15 PM
model
"google/gemini-3.7-flash-20260813"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1298
moderation_latency
(null)
generation_time
11525
tokens_prompt
75
tokens_completion
1622
native_tokens_prompt
67
native_tokens_completion
2972
native_tokens_completion_images
0
native_tokens_reasoning
2224
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
0.01119525
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788443595-qBL5EHpzzhuS1unM2Awz"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788443595-S3F6r7EOiEOE1CqplcPN"
upstream_id
"y3uZavy1OYHSjMcP7eWVqAk"
provider_responses
0
endpoint_id
"dd072012-3274-417d-97e3-e290cba37e31"
id
"y3uZavy1OYHSjMcP7eWVqAk"
is_byok
false
latency
1298
model_permaslug
"google/gemini-3.7-flash-20260813"
provider_name
"Google AI Studio"
status
200
total_cost
0.01119525
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
43.38%
Matches word count
n/a
neededClean
false
words
517
0.00%
Dialogue to Total Word Ratio
Ratio: 90.53%, Deviation: 20.53%
neededClean
false
wordsTotal
528
wordsDialogue
478
21.6892%