NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2GJ1KCCWAN2B7P622V1XY1
Run Details
Model:
DeepSeek V4 Pro
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
10.5s
Cost:
$0.000290
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
20.19%
Matches word count
(details)
27.85%
Dialogue to Total Word Ratio
(details)
24.0175%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 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
The air vents of Chuck E. Cheese smelled aggressively of feet and forgotten pizza. “Tell me the plan again, but this time, with less ambition and more detail,” Molly whispered, her voice a tight wire of irritation. “My knees are soaked in what I hope is root beer.” “It’s elegant, you savage,” Bill’s voice crackled back through the ductwork. “I, disguised as the birthday badger, cause a conga line diversion. You, from above, lower the magnetic claw.” “The claw is a soup ladle on a yo-yo string,” Molly said. “And you’re supposed to be a Chuck E. cousin, not a badger. The kid of the CEO is five. He’s a tiny, vicious critic of woodland authenticity.” “A raccoon, then. A loveable trash prince,” Bill hissed, shifting his weight. The metal groaned beneath him like a dying cow. “The point is, you yoink the mountain of shimmering loot through the ceiling grate while I lead two dozen toddlers in a hypnotic, looping train.” “And the main structural challenge of our meticulously crafted entry point?” “Don’t put your weight on the—” A colossal, tearing screech of metal sheared the sentence in half. Bill vanished, dropping through the ceiling with a comical, brief yelp, followed by the unmistakable sound of a body crashing into a ball pit and a shower of suspended glitter.
created_at
"2026-04-25T14:26:36.272Z"
4/25/2026, 2:26:36 PM
model
"deepseek/deepseek-v4-pro-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1199
moderation_latency
(null)
generation_time
10491
tokens_prompt
77
tokens_completion
352
native_tokens_prompt
67
native_tokens_completion
300
native_tokens_completion_images
(null)
native_tokens_reasoning
0
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"
usage
0.000290145
router
(null)
provider_responses
0
endpoint_id
"d084170b-dd99-4e53-9ae1-b63881f3cfbf"
id
"69300a0d-db8d-47ca-9cb5-8601208edbaa"
is_byok
false
latency
295
model_permaslug
"deepseek/deepseek-v4-pro-20260423"
provider_name
"DeepSeek"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777127196-Xh4Ypy7zoBokBCx4YRkT"
session_id
(null)
api_type
"completions"
id
"gen-1777127196-mYtKzXLLWxD7t3qn5mTa"
upstream_id
"69300a0d-db8d-47ca-9cb5-8601208edbaa"
total_cost
0.000290145
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
20.19%
Matches word count
n/a
neededClean
false
words
220
27.85%
Dialogue to Total Word Ratio
Ratio: 60.63%, Deviation: 10.63%
neededClean
false
wordsTotal
221
wordsDialogue
134
24.0175%