NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCWST7BSSQ27H7QGFRZY1JY
Run Details
Model:
Gemini 3.6 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
22.1s
Cost:
$0.039130
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
99.38%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.6887%
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
"Look at the schematic, Bill. The bounce house provides primary cover," Molly said, tapping a red crayon on a greasy diner napkin. "It is a bouncy castle, Molly. And the target is seven years old today," Bill said, pinching the bridge of his nose. "Are we really doing this? It is a kid's birthday party." "It is the jackpot, Bill! Timmy’s parents are venture capitalists. Do you know what kind of tech is in those shiny wrapped boxes? We are talking high-end tablets, custom gaming consoles, and possibly a pony." "You cannot pocket a pony, Molly." "Not with that attitude, you cannot." Molly leaned across the booth, her eyes wide with unhinged enthusiasm. "Now, the main drop-off is scheduled for fourteen hundred hours, right after the pinata massacre." "The pinata massacre?" "When the little monsters are distracted by cheap sugar, we strike the main gift table." Bill sighed loudly, tracing a finger along the crude crayon drawing of the suburban backyard. "What about security?" "Barnaby the Balloon Guy. High threat level. He carries a concealed electric air pump and has zero tolerance for uninvited guests." "He twisted a sausage dog for me once at a diner. Nice guy, really." "Do not let the inflated dog fool you, Bill! He is a cold-blooded entertainer. If he spots us, he will weave us into terrifying latex shapes. You take out Barnaby." "How am I supposed to do that?" "With the secret weapon." Molly slid a small juice box across the sticky table. "Fruit Punch?" "Organic Apple, four hundred percent daily value of sugar. One sip, and he will be in a total sugar coma by cake time." "That is diabolical. What about the perimeter? Toddlers are unpredictably sticky, Molly." "We wear full tactical velvet. Nothing sticks to velvet. Jam, peanut butter, tears—it all glides right off." "Won't tactical velvet make us look like lounge singers?" "We will look like velvet shadows, Bill! Silent, supple, and extremely well-dressed." Molly unrolled a second napkin, revealing an escape route. "Now, for extraction. We load the loot into the double-wide stroller." "The stroller?" "It has all-terrain wheels and dual cup holders. Perfect for heavy, high-value cargo." "And what if the birthday boy catches us red-handed?" "Timmy is seven. If he confronts you, drop a handful of glitter and whisper that you are the Birthday Ninja. He will be too dazzled to scream." "Glitter? Molly, glitter stays on your skin for weeks! It is a logistical nightmare!" "Sacrifices must be made for the score, Bill." "I still think taking Lego from a child crosses a clear ethical line." "It is not just Lego, Bill. It is the collector's edition space station set. Ten thousand pieces. It belongs in a museum, or our living room." "Fair point. That set is incredible." "So, are you in, or are you going to let a bunch of juice-addled ankle-biters enjoy the finest haul in suburban history?" Bill carefully adjusted his hat and pocketed the juice box. "Pass me the velvet. I am driving the stroller."
created_at
"2026-07-25T15:02:50.869Z"
7/25/2026, 3:02:50 PM
model
"google/gemini-3.6-flash-20260721"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1320
moderation_latency
(null)
generation_time
22017
tokens_prompt
75
tokens_completion
2211
native_tokens_prompt
67
native_tokens_completion
5204
native_tokens_completion_images
0
native_tokens_reasoning
4482
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
0.0391305
router
(null)
provider_responses
0
endpoint_id
"5d6d133d-a953-4781-a835-7ee76f4e1388"
id
"GtBkarr1N_DrjrEPuLvlgAo"
is_byok
false
latency
1320
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-1784991770-vRsI89J2GTkx7f8OJEur"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784991770-4RoLkTQ57PwKcWh5Rnfj"
upstream_id
"GtBkarr1N_DrjrEPuLvlgAo"
total_cost
0.0391305
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
99.38%
Matches word count
n/a
neededClean
false
words
505
0.00%
Dialogue to Total Word Ratio
Ratio: 85.63%, Deviation: 15.63%
neededClean
false
wordsTotal
515
wordsDialogue
441
49.6887%