NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KBSNBSQY2BKTMJY71PVGTF
Run Details
Model:
DeepSeek V4 Pro 0813 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
3m 12s
Cost:
$0.054950
Provider:
StreamLake
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
49.96%
Dialogue to Total Word Ratio
(details)
74.9791%
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 adjusted her earpiece against the heat and peered through the windshield at the inflatable dragon guarding the entrance to Bounce-A-Lot. The air smelled of hot vinyl and failed dreams. Beside her, Bill spread a crayon-drawn floor plan across the dashboard and tapped it with a lollipop stick. "We go in during the piƱata. All adult eyes on the blindfolded six-year-old with a bat." Molly lowered her binoculars and frowned at the drawing. "That's your plan? Wait for a kid to commit assault and then grab the presents?" Bill nodded so earnestly the fake mustache on his upper lip peeled at one corner. "Not just grab. Redistribute. There's a Nintendo Switch shaped like a fire truck and a robotic hamster that farts rainbows." "You want to steal a farting hamster from children." "I want to save those toys from a life of peanut butter and toy-bin neglect. We're providing a better home." Bill folded the floor plan into a paper airplane and tossed it onto the dash. Molly pinched the bridge of her nose. "How do we get inside? We're two adults with criminal records, not party clowns." Bill pulled a red foam nose from the glove box and fixed it over his real nose, which was already suspiciously red. "I'm a clown. You're my glamorous assistant. We'll blend in with the birthday chaos." "You can't juggle." "I can juggle one thing. I'll juggle the smallest present." Molly snatched the foam nose off his face and threw it into the back seat. "What's the distraction? Because if you say hamsters again, I'm leaving." Bill retrieved the nose and dusted it with solemn care. "I rented a magician. Cousin Ralph. He owes me eighty dollars and knows three card tricks." "The man who made a rabbit disappear permanently?" "That's why he's perfect. No rabbit means no animal cruelty charges, just a very confused hat." Bill smiled, exposing a lollipop-stained tongue. Molly looked at the stroller folded in the van's cargo space. "Where do we put forty stolen presents?" Bill followed her gaze. "I brought a double stroller. We pile the boxes in, throw a blanket over them, and push it out like sleep-deprived parents." "What if someone asks to see the baby?" "I'll say the baby is a light sleeper and also a federal witness." Bill licked the lollipop stick and stuck it behind his ear like a cigarette. "That's not a thing." "According to my niece, it is." Molly grabbed the door handle with the resignation of a woman who has hot-wired bouncy castles before. "No eye contact with children. They can smell guilt." Bill stepped onto the asphalt and adjusted his waistband. "Please. I told my nephew the tooth fairy was a hedge fund manager." They walked toward the inflatable dragon. Somewhere inside, a child shrieked with the pure, untamed joy of someone who had found the chocolate fountain. Bill whispered, "That's the sound of our retirement." Molly whispered back, "That's the sound of a kid throwing up." "Same thing."
created_at
"2026-09-03T10:07:26.599Z"
9/3/2026, 10:07:26 AM
model
"deepseek/deepseek-v4-pro-20260813"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2300
moderation_latency
(null)
generation_time
192082
tokens_prompt
75
tokens_completion
15153
native_tokens_prompt
146
native_tokens_completion
16373
native_tokens_completion_images
(null)
native_tokens_reasoning
15695
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
(null)
usage
0.054950181
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788430046-IRQJ0t6NUeYdB1UfEjxw"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788430046-22wCFWuA1w5HktHvcue7"
upstream_id
"req-qm7613-1788430047239056327"
provider_responses
0
endpoint_id
"072c1c23-c069-4242-bb6d-1b30ca5f4eb8"
id
"req-qm7613-1788430047239056327"
is_byok
false
latency
2300
model_permaslug
"deepseek/deepseek-v4-pro-20260813"
provider_name
"StreamLake"
status
200
total_cost
0.054950181
cache_discount
(null)
upstream_inference_cost
0
provider_name
"StreamLake"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
501
49.96%
Dialogue to Total Word Ratio
Ratio: 53.73%, Deviation: 3.73%
neededClean
false
wordsTotal
510
wordsDialogue
274
74.9791%