NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1M1MSX0R6Q4DY38EMYYYCQ5
Run Details
Model:
Hy4 Preview (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
4m 34s
Cost:
$0.050149
Provider:
Tencent
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.97%
Dialogue to Total Word Ratio
(details)
99.9836%
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 pressed her back against the fence, the scent of cupcakes and chaos drifting over the top. Beside her, Bill was busy taping a fake mustache to his upper lip with a strip of duct tape. He had insisted it made him look like an off-duty magician, though Molly thought he looked like a raccoon in formalwear. "This is a terrible idea," Molly said. Bill adjusted his mustache. "This is a brilliant idea. Where else can you steal an entire mountain of gifts while adults are too busy preventing frosting-related injuries to notice?" Molly peered through a gap in the fence. A dozen children sprinted in circles, screaming. Near the patio, a table groaned under the weight of brightly wrapped boxes. A bright banner read HAPPY 5TH BIRTHDAY, TIMMY! The packages ranged from gift cards to a box large enough to hide a pony, and Molly felt the familiar thrill of larceny. "The target is the gift table," Bill whispered. "We go in during the piñata blindness." "The what?" "Blindfolded kid, swinging stick, crowd screaming. It's basically a smokescreen with candy." Molly sighed. "And how do we get past the parents?" Bill held up a clipboard he had stolen from a real estate office. "We're from the Department of Party Safety. Nobody questions a clipboard. I once returned a library book using a clipboard." "That was a pizza menu." "Same energy." They waited. Inside, a woman handed the birthday boy a blindfold. The crowd cheered. Bill nodded. "Now," he said. They slipped through the gate. Their entry point was a side gate, partially blocked by a plastic slide and a determined toddler. Molly kept her head down, marching like a woman on a mission. Bill strolled beside her, clipboard raised, muttering about structural piñata integrity. A father glanced at them, then went back to filming his child with shaking hands. At the gift table, Molly lifted the corner of the tablecloth. "Grab the small ones first," Bill murmured. "High value, low volume. I mean low volume." Molly rolled her eyes and began stacking boxes into a large tote bag printed with a smiling cartoon tooth. She worked quickly, sliding anything shiny or suspiciously light into the bag, because experienced thieves knew that expensive toys came in ridiculous packaging. "Nice bag," Bill said. "Stole it from the craft table," Molly replied. "Irony is free." In the distance, the piñata finally burst, raining candy and causing a brief, beautiful distraction. A girl in a sparkly dress pointed a half-eaten cupcake at them. Molly froze. Bill knelt, mustache slipping. "We're professional present inspectors," he said. "Very serious job. Dangerous, even." The girl narrowed her eyes, then accepted the neon Happy Birthday sticker Bill peeled from his shirt. She shrugged and wandered off. "I hate kids," Molly muttered. "You love loot," Bill reminded her. "Fair point," Molly said. Molly exhaled. "We are never doing this again." Bill grinned and hoisted the bulging tote. "Until next Saturday. I hear Timmy's cousin is turning six."
created_at
"2026-09-03T16:29:16.076Z"
9/3/2026, 4:29:16 PM
model
"tencent/hy4-preview-20260827"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1410
moderation_latency
(null)
generation_time
273307
tokens_prompt
75
tokens_completion
17409
native_tokens_prompt
83
native_tokens_completion
20024
native_tokens_completion_images
(null)
native_tokens_reasoning
19325
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
0.050149246
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788452956-zj9nNaMXIxoIki0HbIyq"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788452956-6aMN15NIQkX7pxBCqgon"
upstream_id
"f685dbf9-cf4c-49d6-96c4-83ff348e1acc"
provider_responses
0
endpoint_id
"ad2a4093-3d2a-41d9-adf1-8b499ac8445d"
id
"f685dbf9-cf4c-49d6-96c4-83ff348e1acc"
is_byok
false
latency
1409
model_permaslug
"tencent/hy4-preview-20260827"
provider_name
"Tencent"
status
200
total_cost
0.050149246
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
499
99.97%
Dialogue to Total Word Ratio
Ratio: 30.08%, Deviation: 0.08%
neededClean
false
wordsTotal
502
wordsDialogue
151
99.9836%