NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCV2MW1NWX3T32G227AB2A0
Run Details
Model:
Gemini 3.6 Flash (Reasoning, Minimal)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
4.7s
Cost:
$0.004376
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0019%
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 surveillance photo was blurry, but there was no mistaking the target: a three-tiered mountain of pastel wrapping paper topped with a giant purple bow. Molly leaned over the kitchen table, adjusting her tactical binoculars. "We strike at fourteen hundred hours, right after the pinata massacre." Bill rubbed his chin, looking at the blueprint of the suburban backyard. "Are you sure about this, Molly? The bouncy castle presents a massive tactical hazard. One wrong bounce and you're airborne over the neighbor's prize-winning begonias." "I’ve factored in the inflation levels, Bill. We slip in during the musical chairs distraction. You grab the lego sets, I handle the cash envelopes from the wealthy grandparents." "What about the target’s mother?" Bill asked, shivering slightly. "I saw her earlier. She's hyped up on espresso and Pinterest. She’s dangerous." "She’s occupied with the juice box distribution. We have a ninety-second window before the clown arrives." "I hate clowns," Bill muttered, pulling a ski mask over his head, only to realize it was a sparkly pink birthday hat. He quickly took it off. "Wait, what's our exit strategy if the toddlers spot us?" Molly pulled two shiny silver cylinders from her trench coat. "Bubbles. High-density, non-toxic, strawberry-scented smoke screen. It will blind them with wonder for at least two minutes." "Brilliant," Bill breathed. "And the dog? The golden retriever?" "I brought bacon-flavored distraction treats," Molly declared, patting her tactical vest pocket. "He’ll be eating out of the palm of my hand." "Literally?" "Literally. Now, review the code names." Bill straightened his spine. "I am Shadow Hawk." "No, you're Uncle Gary. I am Aunt Susan. We are distant relatives from Ohio who lost our invitation." "Right. Uncle Gary. And if someone asks why Uncle Gary is carrying a oversized duffel bag?" "You tell them it contains a very large, very delicate croquet set." Bill stared at her. "At a four-year-old’s birthday party?" "Fine, tell them it's a giant stuffed giraffe. Just don't drop the bag." Molly checked her watch. "It’s time. Deploy the party horns." Bill pulled a plastic noisemaker from his pocket and blew into it. It let out a sad, wheezing squeak. "Intimidating," Molly grunted, sliding down her night-vision goggles despite the blinding afternoon sun. "Let's go steal some juice boxes and wrapped cardboard."
created_at
"2026-07-25T14:32:43.147Z"
7/25/2026, 2:32:43 PM
model
"google/gemini-3.6-flash-20260721"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
588
moderation_latency
(null)
generation_time
4549
tokens_prompt
75
tokens_completion
619
native_tokens_prompt
67
native_tokens_completion
570
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
0.0043755
router
(null)
provider_responses
0
endpoint_id
"5d6d133d-a953-4781-a835-7ee76f4e1388"
id
"C8lkarmsErKPjMcPj_zQ2Q4"
is_byok
false
latency
588
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-1784989963-w7BHZI3FIq9e1dZXMaHl"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784989963-krWFzO12KsCIcUVjhmeY"
upstream_id
"C8lkarmsErKPjMcPj_zQ2Q4"
total_cost
0.0043755
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
379
0.00%
Dialogue to Total Word Ratio
Ratio: 64.27%, Deviation: 14.27%
neededClean
false
wordsTotal
389
wordsDialogue
250
0.0019%