NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCVSYSY4QCF0YK9GZ2DQFWD
Run Details
Model:
Gemini 3.5 Flash Lite (Reasoning, Minimal)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
3.1s
Cost:
$0.001210
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 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 stared at the laminated invitation like it was a classified dossier detailing nuclear launch codes. It was shaped like a smiling unicorn and smelled faintly of artificial strawberry. Bill adjusted his night-vision goggles, which were currently pushed up on his forehead and making his hair stick out like a frightened owl. Are you sure about the intel on this place, Molly? Positive, Bill. Seven-year-olds are terrible witnesses and even worse accountants. They wont notice if the loot is redistributed. The flyer says Princess Sparkles is doing a magic show at three. Princess Sparkles is a twenty-two-year-old drama major named Gary who doesnt want to get pepper-sprayed by two professionals, Molly said, checking her stopwatch. We go in through the bounce house, bypass the juice box barricade, and scoop up every wrapped box under that monstrosity of a pastel pine tree. A bounce house infiltration. Thats tactically degrading. Thats why they wont expect it. Now synchronize your chronometers and check your grappling hooks. Mine is tangled with a spare juice pouch, Bill grumbled, fumbling with a neon pink strap. And why are we wearing tactical turtlenecks? Camouflage is key, Bill. You cant blend in at a suburban cul-de-sac without looking like a brooding European architect. What if we encounter hostile resistance from the birthday boy? Little Timmy? That kid cries if his crusts are left on his peanut butter sandwiches. Ill deploy a handful of glitter if he gets mouthy. Glitter is a war crime, Molly. Its effective. Now, review the extraction route. We load the stolen loot into the Radio Flyer wagon, ditch the unicorn hats, and blend in with the parents lingering by the mimosa station. Brilliant. Just remember, we are after high-value targets only. Legos, remote-control cars, and anything containing chocolate. Leave the homemade macaroni art. Too bulky for resale value, Bill agreed, pulling his ski mask down over his nose. Should we rehearse our cover story? We are distant cousins from the Netherlands here to inspect the bouncy infrastructure. Nobody will question the Dutch. Its foolproof. Let us liberate those presents.
created_at
"2026-07-25T14:45:26.982Z"
7/25/2026, 2:45:26 PM
model
"google/gemini-3.5-flash-lite-20260721"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
324
moderation_latency
(null)
generation_time
2858
tokens_prompt
75
tokens_completion
555
native_tokens_prompt
67
native_tokens_completion
476
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
0.0012101
router
(null)
provider_responses
0
endpoint_id
"6bd8f433-79e1-416c-b407-1772eb796c9a"
id
"B8xkasCWD7e5sOIP9bLC-Qs"
is_byok
false
latency
324
model_permaslug
"google/gemini-3.5-flash-lite-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-1784990726-NI7viJA5ywVaGl43VMRQ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784990726-OjtcR2pvXBsZ9w0h55r7"
upstream_id
"B8xkasCWD7e5sOIP9bLC-Qs"
total_cost
0.0012101
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
345
0.00%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 70.00%
neededClean
false
wordsTotal
356
wordsDialogue
0
0.0000%