NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2FH6VEAP46MN4SBNX5NTQP
Run Details
Model:
DeepSeek V4 Pro (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
23.0s
Cost:
$0.000466
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
20.19%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
28.4888%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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
"Alright, Bill, walk me through the plan one more time." "Simple. I'm Cuddles the Clown. You're Sparkles the Unicorn. We waltz into little Timmy's sixth birthday party, faces hidden by professional-grade face paint, and we make balloon animals until the cake comes out." "And then?" "Then, while the sugar-crazed horde descends on the cake, we stuff every last brightly wrapped present into my oversized clown pants. I've reinforced the crotch with Kevlar and added a quick-release drawstring." "Your clown pants are a duffel bag." "A festive duffel bag of joy, Molly. Focus. The parents will be too busy filming the frosting massacre to notice us waddling out with suspiciously square bulges." "What about the bouncy castle? That's a variable." "It's an alibi, not a variable. If anyone gets nosy, I'll claim Sparkles got her horn stuck in the entrance flap and we have to leave immediately due to mythical creature embarrassment. It's airtight." "It's unhinged. I'm not wearing the horn. Last time it gave me a welt." "Fine, ditch the horn. But you keep the glitter. The blinding, magpie-distraction-grade glitter. That's our exit strategy." "Agreed. So, presents in pants, glitter bomb, flee the scene. You're absolutely sure the hired magician won't be a problem?" "I drugged his rabbit. He's fully occupied with an existential crisis and a very floppy assistant."
created_at
"2026-04-25T14:08:47.583Z"
4/25/2026, 2:08:47 PM
model
"deepseek/deepseek-v4-pro-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1210
moderation_latency
(null)
generation_time
15705
tokens_prompt
77
tokens_completion
556
native_tokens_prompt
67
native_tokens_completion
502
native_tokens_completion_images
(null)
native_tokens_reasoning
175
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
0.000465885
router
(null)
provider_responses
0
endpoint_id
"d084170b-dd99-4e53-9ae1-b63881f3cfbf"
id
"db509d73-0f32-42c6-8a18-953cba16ce89"
is_byok
false
latency
278
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-1777126127-pf4xpvTYPERIjpPrGBgT"
session_id
(null)
api_type
"completions"
id
"gen-1777126127-GF08zy9ze12Ih0LlIMe6"
upstream_id
"db509d73-0f32-42c6-8a18-953cba16ce89"
total_cost
0.000465885
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
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
225
wordsDialogue
225
28.4888%