NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KXK8HW8J37KDWN7WYJW76X
Run Details
Model:
Hy4 Preview (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
3m 47s
Cost:
$0.040193
Provider:
Tencent
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.84%
Dialogue to Total Word Ratio
(details)
99.9213%
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 neighborhood smelled like charcoal and impending HOA fines. The regulation-issue burglar van was parked behind a hydrangea bush, which did little to hide it and even less to make it inconspicuous. Molly adjusted her night-vision goggles, which were actually swimming goggles filled with green food coloring. Bill was wrestling a grappling hook made from a coat hanger and a very optimistic length of dental floss. “Remind me again,” Bill whispered, “why we’re robbing a six-year-old’s birthday party?” “Because,” Molly said, “the intel says the gift table is stacked. We’re talking tablets, scooters, and one of those ride-on cars with the working horn. Also, the guest list has no security, just a mother named Karen who will try to negotiate with us using juice boxes.” “Juice boxes,” Bill repeated. “Truly, the most fearsome weapon in the criminal underworld.” “Just wait until you see the piñata.” Bill peered over the fence. A man in a clown suit was attempting to inflate a giraffe while a toddler supervised with the cold detachment of a tiny dictator. “That’s our mark,” Molly said. “The giraffe?” “No. The presents. Focus, Bill.” “Fine, I’m focused.” Bill rolled his eyes, then tripped over a garden gnome with dignity. They slipped through the side gate, which was unlocked because suburbanites believe fences are decorative. Inside, the backyard looked like a sugar explosion had been curated by a hyperactive event planner. Streamers hung limply in the heat. A table groaned under wrapping paper, gift bags, and a suspiciously expensive drone still in the box. Balloons bobbed on strings like colorful spies, and a banner declared HAPPY BIRTHDAY, CHAD in letters that suggested the cake had already been cried upon. Bill’s eyes widened. “I didn’t know kids got drones. When I was six, I got a rock with googly eyes.” “Times change,” Molly said. “Now grab the duffel and don’t make eye contact with the pony.” “There’s a pony?” “There’s always a pony.” Bill hoisted the empty duffel. It read PROPERTY OF THE WORLD’S OKAYEST GODFATHER, which was accurate but not stealthy. Molly signaled the plan with three coughs, one salute, and an unnecessary jazz hands flourish. Bill nodded, then knocked over a stack of plastic cups. The duffel yawned open, eager to swallow a small fortune in brightly colored wrapping paper and misplaced parental hopes. The clown froze. The toddler pointed. “Intruder!” the tiny dictator shrieked. “Run!” Molly hissed. They bolted for the gift table, because even in retreat, priorities. Bill scooped the drone and a mountain of gift cards. Molly grabbed the ride-on car by the handlebars. The ride-on car’s horn honked once, a tiny, cheerful protest indeed. Then she shouted, “This is why we train!” “We didn’t train!” Bill yelled back. “Then we improvise!” They crashed through the gate, gifts rattling, pony neighing in what Molly would later describe as judgment. Behind them, Karen threw a juice box and missed. “Next time,” Bill panted, “we hit a retirement party. Less running.” “Agreed,” Molly said. “Also, fewer ponies.”
created_at
"2026-09-03T15:18:31.435Z"
9/3/2026, 3:18:31 PM
model
"tencent/hy4-preview-20260827"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1804
moderation_latency
(null)
generation_time
226658
tokens_prompt
75
tokens_completion
13598
native_tokens_prompt
83
native_tokens_completion
16043
native_tokens_completion_images
(null)
native_tokens_reasoning
15285
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
0.040192765
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788448711-nU3XzJZlBUPhUUQsTIX2"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788448711-ctL1LqQCvYW3i5Iipzel"
upstream_id
"d6b7624e-e173-497e-b1e4-a41e000aaddf"
provider_responses
0
endpoint_id
"ad2a4093-3d2a-41d9-adf1-8b499ac8445d"
id
"d6b7624e-e173-497e-b1e4-a41e000aaddf"
is_byok
false
latency
1804
model_permaslug
"tencent/hy4-preview-20260827"
provider_name
"Tencent"
status
200
total_cost
0.040192765
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
99.84%
Dialogue to Total Word Ratio
Ratio: 30.18%, Deviation: 0.18%
neededClean
false
wordsTotal
507
wordsDialogue
153
99.9213%