NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2F522GED3T0GHE07Y54T26
Run Details
Model:
DeepSeek V4 Pro (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
12.0s
Cost:
$0.000402
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
60.28%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
48.5316%
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, walk me through the loot. We talking plush toys or plastic junk I can fence at a car boot sale?" "Bill, you're thinking small and sad. I've scoped the premium haul. The birthday boy's loaded parents have a 'no duplicates' policy. There's a seven-foot giraffe, Bill. It's practically a ladder." "A ladder that squeaks. Look, I don't do whimsy. What's the entry point? I'm not fitting down a bouncy castle chute." "You'll enter as 'Sprinkles the Sad Clown Who Can't Afford Shoes,' which explains your face. I'll be the magician with a leaky glitter cannon providing the diversion. When I pull a rag soaked in lavender oil out of my sleeve and claim it's a dove, the kids will riot. You grab the haul." "Lavender oil? You're going to lightly season the toddlers while I stuff a sack with singing unicorns?" "The singing ones are the decoys. We're after the silent, heavy boxes. The ones that sound like parental guilt and wine subscriptions." "So the plan is I debase myself in a rainbow wig while you commit olfactory assault on minors, all for a giraffe we'll have to bisect to get on the bus." "Don't bisect Gerald. He's the muscle. Now, can you do a convincing honk or will you default to a scream?"
created_at
"2026-04-25T14:02:02.198Z"
4/25/2026, 2:02:02 PM
model
"deepseek/deepseek-v4-pro-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1092
moderation_latency
(null)
generation_time
11950
tokens_prompt
77
tokens_completion
461
native_tokens_prompt
67
native_tokens_completion
428
native_tokens_completion_images
(null)
native_tokens_reasoning
131
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
0.000401505
router
(null)
provider_responses
0
endpoint_id
"d084170b-dd99-4e53-9ae1-b63881f3cfbf"
id
"57c5b7df-7d9e-494e-88b4-8878e7e5b4ec"
is_byok
false
latency
272
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-1777125722-HevL585eNKXRhTX6wnML"
session_id
(null)
api_type
"completions"
id
"gen-1777125722-08YzkOQ0EHYWgzceDYqd"
upstream_id
"57c5b7df-7d9e-494e-88b4-8878e7e5b4ec"
total_cost
0.000401505
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
60.28%
Matches word count
n/a
neededClean
false
words
215
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
216
wordsDialogue
216
48.5316%