NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1HJ90VKPMZTAV3HS3YCYAVB
Run Details
Model:
Gemini 3.8 Flash (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
10.7s
Cost:
$0.010385
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
6.09%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
3.0453%
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
"The blueprint is simple, Bill. Don't look at me like that, it's drawn to scale." Molly tapped the paper with a chewed pencil. "Molly, this is sketched on a grease-stained napkin in purple crayon." "Purple signifies high security. Look closely. At fourteen hundred hours, the primary target enters the backyard." "The target is six years old, Molly." "The target is Timmy Henderson, heir to a mid-sized regional carpet fortune, and his grandparents possess zero financial restraint. The gift table is situated precisely between the hibiscus bush and the bouncy castle. It is prime real estate." "What about perimeter defense?" Bill squinted at the napkin. "Four suburban mothers armed with box wine and passive aggression. Easily neutralized." "How? We agreed no physical violence against civilians." "I won't touch a hair on their highlighted heads. We trigger an ideological distraction. A well-placed rumor about unpasteurized dairy in the artisanal cupcakes, and they'll abandon their posts in panic." "Brilliant. But what about the entertainment? You mentioned a clown on the comms yesterday." "Bongo the Magnificent. Subcontracted talent. Cheap makeup, bad knees, entirely unmotivated." "I despise clowns, Molly. They have dead eyes and squeaky shoes. What if he spots us near the presents?" "If Bongo compromises the perimeter, you deploy the glitter canister." "A glitter bomb? That stuff never comes out of denim, Molly! It is an eternal curse!" "Focus, Bill! Do you want the motorized miniature electric sports car or not?" "Obviously I want the car. It has working headlights and Bluetooth capabilities." "Then pull yourself together. We breach through the southern trellis at fourteen-fifteen, precisely when the piƱata ceremony begins. The blindfolded children wielding wooden bats will cause mass hysteria. We slip behind the plastic slide, bag the haul, and exfiltrate over the cedar fence." "What kind of bags are we using? Traditional canvas duffles?" "Fifty-gallon contractor bags. Heavy gauge. Unscented. We cannot afford to smell like lavender meadow if the neighborhood labradoodle catches our trail." "Smart. Extremely professional. Now, regarding loot division. If there's a Lego Millennium Falcon in that pile, I have dibs." "We split the retail value sixty-forty, leaning toward me, because I acquired the tactical layout." "You pulled the layout from the mother's public Pinterest board!" "That is open-source intelligence gathering, Bill, and it requires immense patience." "You were eating cheese puffs on my couch." "I was working in the field! Now, what did you procure for disguises?" "I brought catering aprons." "Aprons? This is a swashbuckling pirate-themed adventure. Two adults in black aprons look suspicious. Two adults in ruffled shirts and eye patches look like exceptionally committed hired performers." "I look ridiculous in ruffled cuffs." "You will wear the pirate blouse, William, or you forfeit your share of the juice boxes." "Fine. But if a sugar-crazed toddler bites my kneecap, the operation is blown." "Understood. Synchronize watches. We breach the sandbox in five minutes."
created_at
"2026-09-02T17:22:12.476Z"
9/2/2026, 5:22:12 PM
model
"google/gemini-3.8-flash-20260902"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1562
moderation_latency
(null)
generation_time
10690
tokens_prompt
75
tokens_completion
1527
native_tokens_prompt
67
native_tokens_completion
2756
native_tokens_completion_images
0
native_tokens_reasoning
2052
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
0.01038525
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788369732-zU5NCqSsq86WqtOoh4dr"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788369732-3klYAu1EdCt6ScyFaO3m"
upstream_id
"RFuYasX9IfmNjrEPjLDVEA"
provider_responses
0
endpoint_id
"d376f781-516d-42f9-936d-46f19abca6cc"
id
"RFuYasX9IfmNjrEPjLDVEA"
is_byok
false
latency
1562
model_permaslug
"google/gemini-3.8-flash-20260902"
provider_name
"Google AI Studio"
status
200
total_cost
0.01038525
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
6.09%
Matches word count
n/a
neededClean
false
words
477
0.00%
Dialogue to Total Word Ratio
Ratio: 97.33%, Deviation: 27.33%
neededClean
false
wordsTotal
486
wordsDialogue
473
3.0453%