NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM6310FV0YBM9T54YA0MK88
Run Details
Model:
DeepSeek V3 (2025-03-24)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
19.4s
Cost:
$0.000644
Provider:
ModelRun
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 adjusted her fake clown nose and scowled at Bill, who was busy stuffing balloons into a duffel bag. “This has to be the dumbest plan we’ve ever had.” Bill grinned, his oversized rainbow suspenders stretching dangerously. “Nah, last week’s cat-napping scheme was dumber. Who knew so many people would notice their cats were missing?” He tied the duffel shut and tossed it over his shoulder. “This? This is foolproof. Kids’ parties are chaos. Nobody’s gonna notice a couple extra clowns.” Molly tugged at her polka-dotted bowtie. “We’re not even clowns. We’re just wearing clown *accessories*. And since when do clowns carry duffel bags?” “ sad clown?” Bill offered. Then he waved a dismissive hand. “Doesn’t matter. Once the piñata breaks and hyperactive sugared-up toddlers swarm the gifts, we grab the loot and scram. In and out. Five minutes, tops.” Molly peered through the bushes at the backyard party in full swing. Cheap decorations flapped in the breeze, a dozen kids screamed in glee, and a beleaguered-looking father attempted to light candles despite the wind. She sighed. “Fine. But if we get caught, I’m blaming you.” Bill adjusted his neon-green wig. “Deal. Now, here’s the play—you do the ‘distraction.’” “What distraction?” “The one where you pretend to faint.” Molly glared. “That only works if people care enough to come help you.” Bill shrugged. “So faint dramatically. Throw in a convulsion or two. Kids love that.” Before Molly could protest, he shoved her forward. She stumbled into the clearing, arms flailing. The nearest toddler gasped and pointed. “Wow! A *bad* clown!” “No, no, I’m—oh, forget it.” With an exaggerated wheeze, Molly clutched her chest and collapsed face-first onto the grass. The kids stared. A few clapped. One poked her with a stick. Bill seized the moment. With a practiced flourish, he whipped out a candy hose (a repurposed vacuum tube) and started “spraying”angi out jelly beans. “Who wants *ridiculous amounts of sugar*?” he bellowed. The children stampeded toward him like tiny, chaotic wildebeests. Bill backpedaled, luring them away from the gift table. Meanwhile, Molly lay motionless, now being actively ignored. A small dog sniffed her, deemed her uninteresting, and trotted off. Bill tossed a handful of candy into the bushes and sprinted back as the kids dove for the treats. He grabbed Molly’s arm and hauled her upright. “Distraction complete! Go time!” They lunged for the gift pile—only to freeze as a shadow loomed over them. The beleaguered father crossed his arms. “You two aren’t the clowns I hired.” Bill straightened his wig. “Uh. Surprise standby clowns?” The man squinted. Slowly, Molly reached for a wrapped present. The man’s eyes followed her hand. She hesitated. Then, with a sigh, she withdrew her hand and plastered on a smile. “…Magic trick?” Silence. Theumps his face darkened. Bill grabbed Molly’s arm. “RUN.” They bolted, dodging toddlers and abandoned juice boxes, their clown accessories flying off in their wake. Behind them, kids cheered, thinking it was all part of the show. Panting three streets over, Molly doubled over. “Next time,” she wheezed, “we rob a *bank*.” Bill grinned, adjusting his one remaining suspender. “Where’s the fun in that?”
created_at
"2026-02-16T21:36:07.390939+00:00"
2/16/2026, 9:36:07 PM
model
"deepseek/deepseek-chat-v3-0324"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1790
moderation_latency
(null)
generation_time
19370
tokens_prompt
77
tokens_completion
912
native_tokens_prompt
66
native_tokens_completion
819
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
17
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
0.00064383
router
(null)
provider_responses
0
id
"chatcmpl-8202888a02d54554b5c539451876b323"
status
200
is_byok
false
latency
420
endpoint_id
"a3938710-1b99-4a1f-85f4-2494f3f659d3"
provider_name
"ModelRun"
model_permaslug
"deepseek/deepseek-chat-v3-0324"
api_type
"completions"
id
"gen-1771277747-aWvI9R716Or2pSW6EPnA"
upstream_id
"chatcmpl-8202888a02d54554b5c539451876b323"
total_cost
0.00064383
cache_discount
(null)
upstream_inference_cost
0
provider_name
"ModelRun"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0028%
Matches word count
n/a
neededClean
false
words
532
24.0533%
Dialogue to Total Word Ratio
Ratio: 35.34%, Deviation: 5.34%
neededClean
false
wordsTotal
532
wordsDialogue
188
12.0281%