NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1M0SKBHT5FCATVP3RW4V1CM
Run Details
Model:
Hy4 Preview (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
4m 11s
Cost:
$0.048991
Provider:
Tencent
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.72%
Dialogue to Total Word Ratio
(details)
99.8585%
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
The van idled at the curb, a rusted relic that smelled of old fries and bad decisions. Molly studied the house through a gap in the curtains. Pink balloons bobbed on the mailbox. A banner read Happy Birthday Kayden in letters that glittered like a threat. Somewhere inside, a herd of children shrieked with the unholy energy of soda and cake. “This is the target,” Molly said. “Presents in the living room, right?” Bill consulted a crumpled invitation. “According to the intel, yes. Birthday boy wants a robot. His parents bought three.” “Greedy little tyrant,” Molly said. “I respect it.” Bill adjusted his unicorn onesie, which strained across his stomach like a sausage casing. He had rented a pony named Princess Sparklehooves and hidden it behind the garage. Molly’s disguise was simpler: a cheap magician’s kit, a fake mustache, and a rabbit named Steve who had seen things. “Plan is simple,” Bill said. “I cause chaos with the pony. You grab the presents.” Bill patted the stolen golf cart. “Borrowed,” he said. “From a relaxed neighbor.” Molly checked her watch. The party was at peak sugar. In a few minutes the children would gather around the gift table, eyes wide, hands sticky. That was when the loot was most vulnerable. “What about the magician?” she asked. “The real one might show up.” Bill grinned. “I paid him twenty bucks to develop a sudden fear of balloons. He’s at home practicing deep breathing.” “Professional.” “Thank you.” Molly hoisted an empty duffel bag. She slung it over her shoulder, feeling the familiar thrill of a terrible idea gaining momentum. Bill handed her a walkie-talkie shaped like a dinosaur. “Channel two,” he said. “If a toddler tries to arrest you, honk twice.” “That’s your emergency signal?” “It worked at the zoo.” Molly tucked the walkie-talkie into her pocket. The front door of the house opened. A mom with a clipboard scanned the yard, suspicious of the van. Molly slid down in her seat. Bill waved with one glittery hoof. “She’s looking right at us,” Molly hissed. “Smile and blend in,” Bill said. “We’re just two clowns waiting for a parade.” “You’re a unicorn. I’m a magician with a raccoon in my hat.” “Close enough.” Molly peeked again. The mom shrugged and went back inside. The coast was clear, if one ignored the pony eating the host’s hydrangeas. Bill zipped his onesie to the chin. Molly counted the presents in her head: a robot, a drone, and something that beeped. It was almost too easy, which made her nervous. Caution was expensive. Very expensive. Easy jobs had a way of involving orthodontists and lawsuits. “Something wrong?” Bill asked. She shook her head. “Enjoying the calm before the cuffs.” “Showtime,” he said. Molly grabbed her top hat. “Try not to get kicked.” “No promises,” Bill said. “Princess Sparklehooves has opinions.” They climbed out of the van and crept toward the side gate, ready to steal Christmas in July from a six-year-old who would never recover.
created_at
"2026-09-03T16:14:24.689Z"
9/3/2026, 4:14:24 PM
model
"tencent/hy4-preview-20260827"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1918
moderation_latency
(null)
generation_time
251163
tokens_prompt
75
tokens_completion
15977
native_tokens_prompt
83
native_tokens_completion
19561
native_tokens_completion_images
(null)
native_tokens_reasoning
18826
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
0.048991283
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788452064-CxaEToskF6GUBSOLQVO9"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788452064-hld4PfxooIpylWB6REDJ"
upstream_id
"e8c64944-de1f-4a8e-ab4d-2acce5c7b9d8"
provider_responses
0
endpoint_id
"ad2a4093-3d2a-41d9-adf1-8b499ac8445d"
id
"e8c64944-de1f-4a8e-ab4d-2acce5c7b9d8"
is_byok
false
latency
1918
model_permaslug
"tencent/hy4-preview-20260827"
provider_name
"Tencent"
status
200
total_cost
0.048991283
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
99.72%
Dialogue to Total Word Ratio
Ratio: 29.76%, Deviation: 0.24%
neededClean
false
wordsTotal
504
wordsDialogue
150
99.8585%