NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCTTBXH0TXJSEH8EKPRBRK7
Run Details
Model:
Gemini 3.6 Flash (Reasoning, Minimal)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
5.0s
Cost:
$0.004788
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0001%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% 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 heavy oak door of the abandoned warehouse squeaked shut behind Bill. He shook off his damp coat, pulled up a folding chair, and dumped a stack of colorful, glittering pamphlets onto the greasy workbench. Molly leaned over, peering through the dim light of a buzzing fluorescent tube. She frowned. Are those bouncy castles? They are, Bill said, pointing to a pastel pink pamphlet. We strike at fourteen hundred hours. The target is eight years old, named Timmy, and heavily guarded by twenty-two sugar-addicted second graders. Molly pinched the bridge of her nose. Bill, we are professional jewel thieves. We spent three weeks digging a tunnel under the national bank last spring. Which collapsed into the sewer system, Bill reminded her cheerfully. This is different. This is low-risk, high-reward, pure profit. It is a bouncy castle, Bill! And a ball pit, he added. Do not underestimate the ball pit. That is where they store the oversized boxes. Molly sighed, resting her chin in her hands as she looked at a floor plan scribbled entirely in red crayon. What is the layout? Simple, Bill said, tapping the paper. We enter through the back patio under the cover of a piñata distraction. I deploy the smoke bomb. You mean a fog machine. I mean a tactical theatrical fog machine, Bill corrected sharply. Once the smoke hits, the ankle-biters panic. They run for the cake table. That is when you slip past the clown and head straight for the loot pile. What if the clown spots me? Molly asked. Clowns have terrifying spatial awareness. You throw a handful of glitter in his eyes and keep moving, Bill said with absolute deadpan serious intensity. We do not stop for the clown, Molly. Molly rubbed her temples, trying to process the absurdity of her life. Fine. What is the actual payload here? Are we talking video games? Cash envelopes from wealthy grandparents? Even better, Bill whispered, leaning in so close Molly could smell his cheap gas station coffee. Lego sets. Unopened, high-tier Lego sets. The Star Wars ones with three thousand pieces. Molly paused. Her eyes widened slightly. The Millennium Falcon? I saw three distinct Millennium-Falcon-shaped boxes go into that house this morning, Bill confirmed, nodding solemnly. It is a fortune in plastic bricks, Molly. We sell them on the black market, or eBay, whichever has lower seller fees. Molly stared at the crayon map, a slow, criminal smile spreading across her face. Okay, I am in. But I get to wear the tactical balaclava. Agreed, Bill said, pulling out a walkie-talkie. Just remember, once we get the loot, we make a break for the getaway vehicle. The minivan? The minivan, Bill agreed. And for heaven’s sake, do not step on any rogue Legos during the retreat. It will end us.
created_at
"2026-07-25T14:28:11.832Z"
7/25/2026, 2:28:11 PM
model
"google/gemini-3.6-flash-20260721"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
563
moderation_latency
(null)
generation_time
4962
tokens_prompt
75
tokens_completion
703
native_tokens_prompt
67
native_tokens_completion
625
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
0.004788
router
(null)
provider_responses
0
endpoint_id
"5d6d133d-a953-4781-a835-7ee76f4e1388"
id
"-8dkasmtNrLVjMcPx5-fqQY"
is_byok
false
latency
554
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-1784989691-0C2sutt18DQybGsZsbBe"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784989691-0ly2O5TzfzMCy9J4uUGl"
upstream_id
"-8dkasmtNrLVjMcPx5-fqQY"
total_cost
0.004788
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
466
0.00%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 50.00%
neededClean
false
wordsTotal
475
wordsDialogue
0
0.0001%