NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2H1ZD1DQF02BRH6KFB1KT0
Run Details
Model:
DeepSeek V4 Pro (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
10m 11s
Cost:
$0.019965
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.99%
Dialogue to Total Word Ratio
(details)
99.9954%
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
Molly traced a sticky route with a candy cane. "The ball pit is the key. We tunnel under the presents." Bill peered over her shoulder, smelling of bubblegum. "The pit's full of toddlers." "Exactly. Organic cover." "They bite." "Bring bandages. Now, the piñata is a decoy. We rig it with glitter." "Glitter's the herpes of craft supplies." "Precisely. When the blindfold goes on, chaos erupts. We fill the bouncy castle with helium." "Helium?" "Squeaky escapes. No one chases a floating thief." Bill considered this, then nodded. "I'll need a clown nose that honks in D-minor." "Already tuned." She tossed it over with a wink. Bill caught the nose and honked it mournfully. Molly handed him a glitter-coated walkie-talkie. "Channel four. The signal is 'unleash the piñata'." Bill clipped the walkie-talkie to his belt, next to a whoopee cushion. The walkie-talkie crackled with static. He adjusted his oversized, squeaking shoes. "And if I get cornered by a bouncy castle attendant?" Molly cracked her knuckles loudly. "Blame the clown. There's always a clown." They shared a grim look. Just outside, the party tent billowed with the shrieks of excited children. Bill took a deep breath and zipped up the furry mascot suit.
created_at
"2026-04-25T14:35:20.03Z"
4/25/2026, 2:35:20 PM
model
"deepseek/deepseek-v4-pro-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
895
moderation_latency
(null)
generation_time
609510
tokens_prompt
77
tokens_completion
18147
native_tokens_prompt
67
native_tokens_completion
22915
native_tokens_completion_images
(null)
native_tokens_reasoning
22600
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
0.019965195
router
(null)
provider_responses
0
endpoint_id
"d084170b-dd99-4e53-9ae1-b63881f3cfbf"
id
"de2c797f-1ac5-45c8-b741-ffc053cb9f19"
is_byok
false
latency
275
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-1777127720-SeSGFrAG0jtk4M5EX6d9"
session_id
(null)
api_type
"completions"
id
"gen-1777127720-FSRogUNdp6pl1SBH6V2F"
upstream_id
"de2c797f-1ac5-45c8-b741-ffc053cb9f19"
total_cost
0.019965195
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
200
99.99%
Dialogue to Total Word Ratio
Ratio: 49.02%, Deviation: 0.98%
neededClean
false
wordsTotal
204
wordsDialogue
100
99.9954%