NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-03-01KJTZ2YQVW1P0YM5SYJ9TYV9Z
Run Details
Model:
ByteDance Seed 2.0 Mini
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
3m 20s
Cost:
$0.003058
Provider:
Seed
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
99.87%
Dialogue to Total Word Ratio
(details)
49.9362%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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
Behind the dented Maple Street dumpster, Molly leaned against a stack of discarded iridescent party streamers, twirling a slim jim between her fingers like a baton. Inside the fenced backyard, a tinny children’s playlist blared over kid squeals, and a mountain of wrapped gifts glinted under tattered fairy lights. Bill poked his head out, scuffing a scuffed canvas bag of stolen drop cloths against the pavement. “Molly, are you positive the adults are all inside? I saw a dad earlier hauling a giant sheet cake—he’s gonna come out any second with candles.” Molly snickered, tucking the slim jim into her back pocket. “Relax, Bill. I checked the kitchen window five minutes ago. All the chaperones are stuck passing out frosting-covered plastic utensils and bickering over store-bought vs. homemade cupcakes. They’re not leaving that kitchen until someone cries about a missing toy dinosaur.” Bill grunted, kicking a crumpled party favor bag across the driveway. “What if one of the kids catches us? Last time a toddler saw us swiping a garden gnome, he screamed so loud the whole neighborhood thought a raccoon stole his tricycle.” “Please, those kids are too busy doing the floss with their moms to notice a couple of adults creeping around,” Molly said, nodding toward the backyard gate. “What if the dog spots us, though? Last week the neighbor’s golden retriever chased us all the way to the bus stop.” “Wait, what if the whoopee cushion backfires? Last time I sat on one at the grocery store, everyone laughed at me for ten minutes.” Molly snickered again, patting his arm. “Then we’ll blame the birthday boy. Duh. Besides, I brought distractions: the whoopee cushion stash I grabbed from the dollar store clearance rack.” Bill’s face paled. “You brought whoopee cushions? That’s a middle school prank, not a heist tool!” “Okay, first of all, heist tools don’t come with a refund policy at dollar stores,” Molly retorted, digging a crumpled pack of neon whoopee cushions out of her jacket pocket. “Second, the library book heist last month? That whoopee cushion under the librarian’s chair got us past her before she even heard us sneak out with the rare graphic novel.” “Fine, but if we get caught, I’m blaming you entirely,” Bill huffed, grabbing a pair of nitrile gloves from his bag. “Also, what if all these presents are just stuffed animals and coloring books? We’re gonna leave with $20 max if we’re lucky.” “Have you seen the Facebook post Mrs. Henderson made last night?” Molly asked, leaning in to tap Bill’s shoulder. “She said she got a $50 Target gift card for her daughter’s art supplies, plus a bunch of vintage Lego sets from her cousin. Those Legos alone will pay for our rent for two weeks. The coloring books? Flip ’em at the flea market this weekend. Stuffed sloths? My cousin sold a similar one for $25 last month.” Bill sighed, pulling the bag of drop cloths over one arm. “Fair. But if we’re going down, I get first dibs on the leftover piñata candy.” “Deal,” Molly said, shoving one whoopee cushion under the backyard gate’s rusted latch, wedging it open just enough to slip through. “Now, the playlist just hit the ‘Baby Shark’ dance breakdown, so everyone’s glued to the backyard. Let’s move—before that dad finishes lighting the candles and brings the cake out here.”
created_at
"2026-03-03T23:03:47.72Z"
3/3/2026, 11:03:47 PM
model
"bytedance-seed/seed-2.0-mini-20260224"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3431
moderation_latency
(null)
generation_time
200389
tokens_prompt
77
tokens_completion
8184
native_tokens_prompt
101
native_tokens_completion
7619
native_tokens_completion_images
(null)
native_tokens_reasoning
6837
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
0.0030577
router
(null)
provider_responses
0
endpoint_id
"9265aa87-9657-4d40-b9d7-2a70318e57d9"
id
"021772579027835ef9a07adf34cf0fac1320e81624057e0ee6617"
is_byok
false
latency
3431
model_permaslug
"bytedance-seed/seed-2.0-mini-20260224"
provider_name
"Seed"
status
200
api_type
"completions"
id
"gen-1772579027-N7VsnC1qQ1megOmSe2AB"
upstream_id
"021772579027835ef9a07adf34cf0fac1320e81624057e0ee6617"
total_cost
0.0030577
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
561
99.87%
Dialogue to Total Word Ratio
Ratio: 70.16%, Deviation: 0.16%
neededClean
false
wordsTotal
563
wordsDialogue
395
49.9362%