NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCSF14W1X3YEMVY0V4ESC10
Run Details
Model:
Gemini 3.6 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
59.5s
Cost:
$0.119898
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
90.48%
Matches word count
(details)
94.41%
Dialogue to Total Word Ratio
(details)
92.4493%
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
Molly spread the strategic blueprint across the greasy hood of their rusted getaway van. It was drawn entirely in purple crayon on a grease-stained placemat from Barnaby’s Pizza Palace, but to a professional criminal, a map was a map. Bill leaned over her shoulder, chewing loudly on a handful of stolen cheese puffs, his bewildered face illuminated only by the flickering yellow streetlight. He adjusted his oversized black tactical balaclava very nervously. "Look here, Bill. This bouncy castle is the drop zone. Timmy turns seven at fourteen hundred hours. By fourteen fifteen, the mark will be distracted by the magician." She pointed her dirty index finger at a poorly drawn dragon on the paper. "Is it Barnaby the Bumbling Clown? Because he does that trick with the fake dove that really messes with my brain." Bill shuddered at the memory of past birthday parties. "Focus! It’s a balloon artist named Captain Twist. We have a three minute window while he bends a latex dachshund to slip past the patio doors." Molly slapped his hand away from the pizza placemat and glared at him. "I see the loot, Molly. It’s a mountain. Wrapped in shiny foil with big red bows. I think I spot a Nintendo." Bill pulled a pair of plastic opera glasses from his yellow trench coat and squinted toward the suburban backyard across the quiet neighborhood street. "We aren't here for video games, you amateur. We’re sweeping the entire table. Board games, Lego sets, envelope money—everything goes into the burlap sacks." She tightened her black leather gloves with extremely aggressive precision. "What about the juice boxes? My blood sugar dips around two." Bill scratched his chin thoughtfully. "No juice boxes! We stick to the payload. You grab the left side of the table, I grab the right. If any young kid spots us, what do you do?" Molly exhaled slowly through her nose. "Throw juice in their eyes?" Bill held up a tiny plastic straw like a dagger. "No, you idiot! You drop to one knee, compliment their light up sneakers, and tell them you’re the party entertainment. Then we extract." Molly pulled a heavily modified pink plastic bubble blower from her leather holster. "Is that the Bubble-O-Matic 3000? Molly, that’s tactical genius. They’ll be too dazzled by bubbles to notice us liquidating the gift cache." He touched the shiny toy barrel reverently. "Precisely. We strike when the cake arrives. Sugar rush creates chaos. Chaos creates opportunity." She checked her digital watch and nodded with grim, quiet enthusiasm. "Can we steal a slice of ice cream cake? Mint chip?" His stomach rumbled loudly inside the cramped, metallic van. "If it’s on the table, it’s fair game. The toddler menace awaits." She pulled her black ski mask down, leaving only her eyes visible. Bill fastened a festive, pointed paper party hat onto his oversized head. "For the ultimate score, partner. Let’s rob these rugrats blind." He slapped the van's side door open, ready to infiltrate the suburban backyard birthday party of seven year old Timmy Jenkins without mercy.
created_at
"2026-07-25T14:04:31.786Z"
7/25/2026, 2:04:31 PM
model
"google/gemini-3.6-flash-20260721"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1373
moderation_latency
(null)
generation_time
59398
tokens_prompt
75
tokens_completion
5004
native_tokens_prompt
67
native_tokens_completion
15973
native_tokens_completion_images
0
native_tokens_reasoning
15283
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
0.119898
router
(null)
provider_responses
0
endpoint_id
"5d6d133d-a953-4781-a835-7ee76f4e1388"
id
"b8Jkas_9M4LGjMcP1-XhsQ0"
is_byok
false
latency
1373
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-1784988271-FmlIpV1kR7xweyH1xpz6"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784988271-H7uhPqCyJ9fWKaQ7ctFj"
upstream_id
"b8Jkas_9M4LGjMcP1-XhsQ0"
total_cost
0.119898
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
90.48%
Matches word count
n/a
neededClean
false
words
510
94.41%
Dialogue to Total Word Ratio
Ratio: 51.07%, Deviation: 1.07%
neededClean
false
wordsTotal
513
wordsDialogue
262
92.4493%