NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY982W4E9JZNWKS0JS903VZ
Run Details
Model:
ByteDance Seed 1.6 Flash
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
37.8s
Cost:
$0.001787
Provider:
Seed
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
46.73%
Dialogue to Total Word Ratio
(details)
23.3652%
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
Bill had a neon-yellow juice box sleeve stuck to his wrist (stolen, not intentional). “Smart?” he mumbled, craning around the oak. “Last piñata job: plate hat.” Backyard chaos—toddler wailed over a sippy cup, six-year-old flicked a lollipop stick. Donkey piñata creaked. Molly, on a stolen folding chair, adjusted a store-bought earpiece. “Kids’ parties: weakest at the gift table. Ripped paper, no cameras, parents busy with dog poop.” Bill scowled, ski mask sliding down (stolen from a dumb bank—eyes look sad blueberries). “Starlight Unicorn Robot? Lights up, sings, comes with Sparkle Pony. Worth jello? But kid tears—” “Easier than adult. Watch.” She pointed to rainbow presents, a few torn. “Clock’s ticking. Donkey’ll explode. Sugar rushes. Take Unicorn, dino kit. Dragon crayons—look important. Kid’s face? Priceless.” Bill fumbled his mask. “What if they see us? Lollipop kid’s too close—smelling the bushes.” Molly squinted. Six-year-old tilted his head, confused. “Checking if we’re piñata. Ignore. Distraction.” She pulled a whoopee cushion (stolen party supply). “Squeeze when I say ‘now’—he’ll pee. Kids lose it over farts. Study once.” Bill stared. “Fart noise? Sounds like—” “A fart. Kids go nuts. Trust me.” Lawn dart tinkled, toddler wailed. Molly checked her phone—1:17. “Go. Grab Unicorn, dino, crayons. Gift table. Bill?” “Yeah?” “Don’t touch jello. Mint, stains gloves—neon ‘World’s Okayest Thief’—everyone’ll know.” “Gloves are fine!” “They’re not. Saw you steal ‘em from the craft store.” “Shut up and hand the whoopee cushion.” Six-year-old stepped closer. Molly squeezed: honk. Kid bolted, cackling. “Told you.” Bill inched toward gifts, elbow in ribbon. “Move over, gorilla.” She tossed scissors. “Taped like a mummy—no cheap knockoffs.” Twin stopped wailing. Mom hurried over, wiping toddler. “Noise okay?” “Squirrel, dear.” Bill tugged Unicorn—beeped, garbled “Twinkle Twinkle.” Mom glanced over, back to kids. “Weird.” Molly grabbed dino, crayons, animal crackers. “Got it—nod.” “Let’s book it before piñata hits.” Bill stood, sleeve flapping. “Need bigger distraction?” “Nah. Fart noise enough. Kids remember forever. Therapists later.” Piñata cracked, candy rained. Kids dived. Molly and Bill slipped, shoulders brushing the fence. “Not bad,” Bill said. “Skip balloon decor? Looked like an airheaded sponge.” “Absolutely—if you hold the unicorn. Mine. Jello’s yours.” Bill paused. “Jello with extra berries?” “Hypothetically. Come on—before the balloon kid returns.” Neighbor Marge’s voice crackled: “Molly? Dog ate key—groom him first.” “Later, Marge. Priorities.” They melted into the street, unicorn beeping, dino clinking. Somewhere, a kid screamed, “MOM, SPARKLES!” Bill grinned. “Best heist.” Molly nodded. “Circus next—lions nothing.”
created_at
"2026-02-20T19:43:23.026Z"
2/20/2026, 7:43:23 PM
model
"bytedance-seed/seed-1.6-flash-20250625"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
628
moderation_latency
(null)
generation_time
37604
tokens_prompt
77
tokens_completion
5934
native_tokens_prompt
149
native_tokens_completion
5920
native_tokens_completion_images
(null)
native_tokens_reasoning
5118
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
0.001787175
router
(null)
provider_responses
0
endpoint_id
"e01a1227-001b-44d4-a376-90850668fa84"
id
"0217716166033769601394fd5171fb187b1a439528361d3a6412c"
is_byok
false
latency
514
model_permaslug
"bytedance-seed/seed-1.6-flash-20250625"
provider_name
"Seed"
status
200
api_type
"completions"
id
"gen-1771616603-4CfCEBgmOFp2R72D5fvF"
upstream_id
"0217716166033769601394fd5171fb187b1a439528361d3a6412c"
total_cost
0.001787175
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Seed"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
423
46.73%
Dialogue to Total Word Ratio
Ratio: 53.90%, Deviation: 3.90%
neededClean
false
wordsTotal
423
wordsDialogue
228
23.3652%