NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHYA32T0QDDTQPKWB6MV47Y5
Run Details
Model:
ByteDance Seed 1.6 Flash
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
15.6s
Cost:
$0.000599
Provider:
Seed
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 propped her chin on her knuckles, eyeing the half-eaten birthday cake on Bill’s kitchen table. Frosting smudged her cheek—she’d stolen a taste before he’d noticed. “You’re staring again,” she said, not looking up from the spreadsheet scrawled on the back of last month’s utility bill. “Relax. The party’s Saturday. 3 p.m. at Mrs. Hargrove’s bakery. Kids, balloons, a piñata that’ll probably collapse halfway through.” Bill flicked a crumb off his sleeve. “A piñata’s not the problem. It’s the *guards*. Dozens of them. Parents. Teachers. That one mom from last week who recognized me from the grocery store because I’d tried to swap her apples for turnips. She held up a list.” He mimed a stern voice: “‘Do not trust the man in the blue bandana. He steals *strawberries*.’” Molly snorted. “Relax. I’ve got a distraction. Mrs. Hargrove’s cousin’s kid is throwing the party—Lila, she’s eight. I texted her yesterday. Told her I’m a clown named Mr. Balloon-Boy. Said I’ll ‘magically’ make the piñata stay upright if she gives me front-row seats. She’ll be too busy gushing to notice when I ‘accidentally’ knock over the presents.” Bill glanced at the spreadsheet. “Target: 27 presents. According to Lila’s Instagram, there’s a Lego castle, a ‘princess tea set that *actually* pours,’ and a remote-control dinosaur that walks on its own. Are you *kidding*? That dino’s probably a TikTok special.” “To *us*, it’s gold,” Molly said. “The Lego castle’s missing the lion minifig—we can sell it separately. The tea set? Rich people love ‘vintage’ knockoffs. And the dino? Bill, this thing has a ‘steam’ function. That’s ‘magic’ to a collector. Besides—” she held up a finger “—the real prize is in the wrapping. Lila’s grandma sent a gift. Said it’s ‘important.’ Probably a watch or a necklace, but—” “Or a jar of pennies,” Bill muttered, popping a chocolate chip into his mouth. Molly flicked a fork at him. “Or not. But we’re not here to be picky. And for the love of God, don’t call it ‘stealing.’ It’s a *diversion*. A *temporary reallocation*.” Bill sighed, but he was smiling. “You’re impossible. What about the alarm? Mrs. Hargrove installed that new motion sensor system after last year’s cookie heist.” Molly grinned. “I disabled it. Last week. While she was at the grocery store, ‘helping’ her pick out wine. She *hates* wine. Now the sensor’s just a ‘fancy phone charger’—Lila plays with it sometimes.” Bill stood, knocking over a bag of party hats. “What about the kids? What if one of them sees me? What if they *cry*?” Molly stood too, grabbing his wrist. “They’ll cry. Then their parents will rush over, and Mrs. Hargrove will be too busy hugging them to notice I’m loading a duffel bag with wrapping paper. You’ll be the ‘worried uncle’ who ‘stole the hat.’” She pressed a red party hat into his hands. “Put it on. It’ll make you look ‘harmless.’” Bill stared at the hat. “I’m a grown man. Wearing a party hat.” “Exactly. Kids think grown men in party hats are fairies. They’ll ignore you. Besides—” she winked “—you’re *great* with kids. Remember when you conned that little girl out of her lollipop by promising to teach her to juggle?” Bill winced. “That was a *moment*. And I returned it. Eventually.” “Exactly. Now c’mon. We need to map the layout. Mrs. Hargrove’s bakery has a back hallway—” “Molly—” “—and Lila’s presents are under a table near the window. If we move fast, we can be in and out before the piñata breaks. And Bill? If you mess up, I’m blaming the dino.” Bill sighed, but he stuck the hat on his head. The brim flopped over his eyes. “Fine. But if a parent asks me if I’m ‘working,’ I’m telling them I’m ‘on strike.’” Molly laughed. “Deal. Now let’s steal some toys. And maybe a cupcake. For the road.” She grabbed the duffel bag, slung it over her shoulder, and headed for the door. Bill followed, tripping over a streamer he’d forgotten was there. “Remind me why we do this again?” he said, brushing confetti off his pants. Molly grinned. “Because loot. And because Mrs. Hargrove deserves to worry about her wine. *Finally*.” They slipped out, the door clicking shut behind them. Somewhere in the distance, a kid yelled, “Mr. Balloon-Boy’s here!” “Perfect,” Molly said. “Hit it.” Bill sighed, but he pulled on his gloves. “Can we get tacos afterward? The ones with the extra guac.” Molly paused. “You *always* ask for guac.” “Guac is non-negotiable,” Bill said, grinning. “And also, distraction. For the parents. When they ask where we got the presents. I’ll be ‘explaining’ guac to them.” Molly laughed. “Let’s go steal stuff. And eat tacos. In that order.” They vanished into the morning, the party still hours away. And somewhere, a piñata waited, unaware.
created_at
"2026-02-20T19:58:07.692Z"
2/20/2026, 7:58:07 PM
model
"bytedance-seed/seed-1.6-flash-20250625"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
448
moderation_latency
(null)
generation_time
15401
tokens_prompt
77
tokens_completion
2097
native_tokens_prompt
149
native_tokens_completion
1958
native_tokens_completion_images
(null)
native_tokens_reasoning
665
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
0.000598575
router
(null)
provider_responses
0
endpoint_id
"e01a1227-001b-44d4-a376-90850668fa84"
id
"021771617487837f2dfc9d17a425ad03ee233374d2636611a0279"
is_byok
false
latency
303
model_permaslug
"bytedance-seed/seed-1.6-flash-20250625"
provider_name
"Seed"
status
200
api_type
"completions"
id
"gen-1771617487-JAtbftp77dfYEq4L1qAg"
upstream_id
"021771617487837f2dfc9d17a425ad03ee233374d2636611a0279"
total_cost
0.000598575
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
809
0.00%
Dialogue to Total Word Ratio
Ratio: 71.82%, Deviation: 41.82%
neededClean
false
wordsTotal
809
wordsDialogue
581
0.0000%