NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2DW1RQY1FE19M8CM3ATM7M
Run Details
Model:
DeepSeek V4 Pro (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
35.4s
Cost:
$0.001066
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 peered through a pair of rhinestone-encrusted binoculars, the kind that came free with a children’s spy kit, and adjusted the focus with a delicate twist of her finger. The target was a pastel-coloured community hall, its windows fogged with the steam of tiny, overexcited bodies. A banner above the door read “Happy 7th Birthday, Archibald!” in glittering, uneven letters. Bill, crammed into the driver’s seat of a minivan that still smelled faintly of crayons and regret, was trying to fold a city map into the shape of a swan. He gave up and let it spring open, crackling like a nest of cellophane. “I still say we just walk in as the extra entertainment. I can juggle. Sort of.” “You can’t juggle, Bill. You dropped a raw egg on my mother’s Persian rug at Easter, and that was just one egg.” Molly didn’t lower the binoculars. “Our intel says there’s a professional magician in there. That means security will be tight. We’re not dealing with an amateur clown who makes balloon poodles. This is a man who makes things disappear for a living. He’ll spot us.” Bill gestured to his outfit, a hastily assembled approximation of a courier’s uniform, complete with a name badge that read “Stan.” “But the decoy boxes. They’re genius. Three large crates marked ‘Live Emperor Scorpions – Handle With Utterly No Care Whatsoever.’ No kid wants that. They’ll clear the room, and we scoop the gift table.” “And when the birthday boy’s mother asks to see the scorpions?” Molly swivelled in her seat, her expression a masterpiece of weary patience. “What then, Stan?” “Improvise,” Bill said, his voice scaling up an octave. “A mime act. I’m a mime cursed by a wizard to be trapped in a box. A distressed courier-mime. It’s the circuit’s next big thing.” Molly returned to her surveillance, the binoculars catching the distorted, wobbly shape of a child wearing a princess dress sprinting past a window. A small, sticky handprint smeared itself across the glass from the inside. “Fine. The mime is plan C. Plan A is still the piñata. You’re inside.” “I’m not getting inside the piñata again. Last time, the birthday boy was a southpaw with a cricket bat and a vendetta. My ribs were confetti-coloured for a fortnight.” Bill rubbed his side subconsciously. “Why can’t you ever be the inside man, or woman, or inanimate paper-mâché donkey?” “Because I’m the wheelman, Bill. The elegant, unsung hero who ensures our escape route is paved with smooth asphalt and zero police cruisers. Also, I simply refuse to be beaten with a stick by children hopped up on cake frosting.” She tapped the van’s dashboard clock. “We breach at the magic show’s finale, the moment of peak distraction. You’ll enter through the kitchen, commandeer the birthday child’s pedal-powered miniature car as a getaway vehicle across the hall, and load the loot into the piñata’s hollow carcass, which you’ll be wearing for armour. I’ll be parked by the bouncy castle’s emergency exit valve.” Bill stared at her, a profound admiration dawning in his eyes. “The pedal car… for acceleration?” “For the look of it, Bill. A grown man in a papier-mâché donkey, speeding away in a pink plastic coupe with a bow on the hood. They won’t chase you. They’ll be too busy questioning the stability of their reality. Now, synchronise your wristwatch with mine.” She held up a plastic, digital watch with a cartoon cat on its face. Bill looked at his own wrist, bare but for a smudge of engine grease. He sighed, the sound of a man who has long since surrendered to the glorious, chaotic logic of his partner. “My wristwatch is in the glovebox. Next to the party blowers and the emergency glitter.”
created_at
"2026-04-25T13:39:38.395Z"
4/25/2026, 1:39:38 PM
model
"deepseek/deepseek-v4-pro-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
912
moderation_latency
(null)
generation_time
34930
tokens_prompt
77
tokens_completion
1307
native_tokens_prompt
67
native_tokens_completion
1192
native_tokens_completion_images
(null)
native_tokens_reasoning
293
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"
usage
0.001066185
router
(null)
provider_responses
0
endpoint_id
"d084170b-dd99-4e53-9ae1-b63881f3cfbf"
id
"bc2c0486-d48f-4a39-aaf6-13a6f14d9e40"
is_byok
false
latency
281
model_permaslug
"deepseek/deepseek-v4-pro-20260423"
provider_name
"DeepSeek"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777124378-Up99xzGQEpwq2j8qqBhC"
session_id
(null)
api_type
"completions"
id
"gen-1777124378-RfH961gWFbUEPcnEofXu"
upstream_id
"bc2c0486-d48f-4a39-aaf6-13a6f14d9e40"
total_cost
0.001066185
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
625
0.00%
Dialogue to Total Word Ratio
Ratio: 60.13%, Deviation: 30.13%
neededClean
false
wordsTotal
632
wordsDialogue
380
0.0000%