NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1M0S6BGK953PNA1PRY5EQH9
Run Details
Model:
Hy4 Preview (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
3m 59s
Cost:
$0.048871
Provider:
Tencent
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.81%
Dialogue to Total Word Ratio
(details)
99.9032%
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
Molly spread the stolen party blueprint across the hood of their getaway minivan, weighting the corners with juice boxes. The Pemberton house glowed like a sugar-fueled beacon. Through the window they could see a bounce house, a long table groaning under presents, and a banner reading HAPPY 6TH BIRTHDAY, TIMMY. The street was quiet except for the distant squeal of a recorder being murdered by a tone-deaf child. Molly’s getaway vehicle, a beige minivan named Steve, idled at the curb with a dented sliding door and a bumper sticker that read HONK IF YOU LOVE NAPS. She had boosted the blueprint from the party planner’s car, bribed a caterer for the guest list, and spent three days practicing a disappearing act that mostly involved dropping scarves. “Tell me again why we’re robbing a six-year-old,” Bill said, adjusting his itchy clown wig. “Jewelry doesn’t scream when you take it.” Molly tapped the blueprint. “Jewelry stores have cameras, guards, and grown-ups. This party has thirty kids high on cake, two exhausted parents, and a piñata. The only security is a retriever named Mr. Wiggles.” Bill groaned. “I’m allergic to dogs. And children. Mostly children.” “You’re fine,” Molly said. “You’re Bobo the Clown. I’m Miss Molly the Magnificent. We perform, the kids gather, and while they’re dazzled, we swap the presents for empty boxes.” “Empty boxes?” Bill echoed. “We’re thieves, not Amazon.” Molly grinned. “Exactly. Nobody questions a clown carrying a box. By the time they unwrap an empty box, we’ll be in Toledo.” Bill studied the blueprint. “What about the dog?” The golden retriever lay on the porch, snoring like a chainsaw wearing a sweater. One ear twitched. A half-eaten cupcake sat glued to his fur, a silent warning that sugar made him unpredictable. “Mr. Wiggles is bribable,” Molly said. “I brought bacon. You brought the world’s saddest balloon animals. Either he loves us or he eats us. It’s a win-win.” Bill slumped. “And if a kid cries?” Molly handed him a whistle. “You juggle.” “I can’t juggle,” Bill said. “Then juggle worse,” Molly said. “Improvise. Our window opens at four when Timmy’s dad breaks out the punch and the piñata explodes. Sugar chaos is our friend.” “Four o’clock,” Bill muttered. “We’re stealing from a birthday party during a candy riot.” Molly checked her watch. “We’re not stealing. We’re entertaining. I hacked the guest list. We waltz in as help, wow the crowd, then leave with every present and nobody the wiser.” Bill adjusted his oversized shoes. “Last question. If we get caught, do we tell them it’s a surprise party for the house?” Molly slung a bag of empty boxes over her shoulder. “Bill, if we get caught, I’ll tell them you’re a rogue clown with a serious gift-card addiction. Now put on your smile. It’s showtime.” She rang the doorbell. Somewhere inside, thirty children screamed with joy. Bill’s fake smile wobbled. Molly’s didn’t. If the plan worked, they’d be legends. If it failed, at least the kids would remember Bobo.
created_at
"2026-09-03T16:14:11.319Z"
9/3/2026, 4:14:11 PM
model
"tencent/hy4-preview-20260827"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2567
moderation_latency
(null)
generation_time
238992
tokens_prompt
75
tokens_completion
15887
native_tokens_prompt
83
native_tokens_completion
19513
native_tokens_completion_images
(null)
native_tokens_reasoning
18724
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
service_tier
(null)
usage
0.048871235
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788452051-Qmb13xczNsmd8sCBbj0r"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788452051-5ht2JfWTKVyMv422ahY5"
upstream_id
"8a34b728-c7a2-48d1-ba0e-d5a760c746cc"
provider_responses
0
endpoint_id
"ad2a4093-3d2a-41d9-adf1-8b499ac8445d"
id
"8a34b728-c7a2-48d1-ba0e-d5a760c746cc"
is_byok
false
latency
2567
model_permaslug
"tencent/hy4-preview-20260827"
provider_name
"Tencent"
status
200
total_cost
0.048871235
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Tencent"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
500
99.81%
Dialogue to Total Word Ratio
Ratio: 50.20%, Deviation: 0.20%
neededClean
false
wordsTotal
508
wordsDialogue
255
99.9032%