NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5Z3Y04G2JWA2EMRXVCBJB
Run Details
Model:
DeepSeek V3.1
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
51.3s
Cost:
$0.000597
Provider:
SambaNova
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
The air in the borrowed van was thick with the scent of stale coffee and desperation. Molly tapped a glossy, brightly colored invitation against the steering wheel, her brow furrowed in concentration. "The target is a five-year-old named Bartholomew. Birthday party. Backyard of the Henderson residence. Three PM sharp." Bill, a large man crammed into the passenger seat, snorted. "Bartholomew? Who names a kid Bartholomew? His friends probably call him 'Barf.'" He squinted at the detailed schematic Molly had drawn on a napkin. It featured a stick-figure dog labelled ‘Baxter – distract with beef jerky’ and a series of swirling arrows leading to a ‘Presents Pile X’. "Focus, Bill. The objective is the loot. High-value items: plasticware, battery-operated annoyances, and, if we're lucky, cash-filled envelopes from forgetful grandparents." "This feels wrong," Bill muttered, scratching his chin. "I mean, we lifted the Crown Jewels of Prague. We swapped the real Mona Lisa with a paint-by-numbers version and nobody noticed for a week. And now we're casing a kid's party for a talking teddy bear?" Molly shot him a look that could curdle milk. "The Crown Jewels had laser grids. The Louvre had pressure-sensitive floors and a small army of guards. This..." she jabbed the invitation, "...has a piñata and a dad who will be three beers deep by four o'clock. It's the perfect score. Low risk, high reward." "What's our exit strategy? Tricycle? A getaway Big Wheel?" "The wheelie bin," Molly said without a hint of irony. "We load the goods into the Henderson's own municipal-issued rubbish bin and roll it nonchalantly down the street. It's the ultimate camouflage. Who's gonna question someone taking out the trash?" Bill stared into the middle distance, a pained expression on his face. "I can see the headlines. 'Master Thief Nabbed by Six-Year-Old with a Net.' My reputation. My standing in the international criminal community..." "Your standing is currently 'unemployed and owes me forty dollars'," Molly countered. "Now, your role is crucial. You'll be dressed as the party clown." Bill’s head snapped towards her. "Absolutely not. The Geneva Convention prohibits cruel and unusual punishment. I have coulrophobia." "You have a fear of bank alarms and gainful employment. You'll be fine. It's a disguise. You’ll be ‘Pogo the Pal’. You create a diversion. A big one. Maybe you start making balloon animals that look… sinister." "Sinister how?" "I don't know, Bill, use your imagination! A poodle with too many legs. A giraffe with a tear in its eye. Something that will captivate and mildly unsettle both children and parents. While they're staring at your existential nightmare in latex, I'll be quietly filling a sack with educational toys." Bill slumped in his seat, defeated. "I just… I never thought my life of crime would lead me to stealing from toddlers. The moral ambiguity is staggering." Molly put a hand on his shoulder. "Think of it not as stealing presents," she said, her voice softening into a conspiratorial whisper. "Think of it as… early-life lessons in disappointment. We’re not thieves, Bill. We’re cynical life coaches." He sighed, a long, weary sound that fogged up the window. "Do I at least get a squirting flower?" "Top of the line," Molly promised, starting the van. "Now, let's go. We need to case the party supply store for a wig that doesn't look like it died of fright."
created_at
"2026-02-16T21:34:31.073331+00:00"
2/16/2026, 9:34:31 PM
model
"deepseek/deepseek-chat-v3.1"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1682
moderation_latency
(null)
generation_time
51229
tokens_prompt
95
tokens_completion
892
native_tokens_prompt
67
native_tokens_completion
783
native_tokens_completion_images
(null)
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)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.0005973
router
(null)
provider_responses
0
id
"ec296822-9fb0-4cb8-8f04-758e7c51c162"
status
200
is_byok
false
latency
1681
endpoint_id
"1dda4bb9-c626-4816-bb54-99665323fef0"
provider_name
"SambaNova"
model_permaslug
"deepseek/deepseek-chat-v3.1"
api_type
"completions"
id
"gen-1771277619-YLKRkYfhjDN11tuQbGhU"
upstream_id
"ec296822-9fb0-4cb8-8f04-758e7c51c162"
total_cost
0.0005973
cache_discount
(null)
upstream_inference_cost
0
provider_name
"SambaNova"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0000%
Matches word count
n/a
neededClean
false
words
569
0.0000%
Dialogue to Total Word Ratio
Ratio: 74.87%, Deviation: 44.87%
neededClean
false
wordsTotal
569
wordsDialogue
426
0.0000%