NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2H83EAMZ21KCEV3VFHCP83
Run Details
Model:
DeepSeek V4 Flash
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
14.8s
Cost:
$0.000092
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
99.38%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
68.0824%
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
“Bill, you sure this is the right address? This looks like a clown’s storage unit.” “Molly, I’ve scoped it. The party’s in the backyard. The gifts are piled by the inflatable unicorn.” “A unicorn? Great. So my heist costume involves dodging a six-year-old on a sugar high named Princess Sparkle.” “Focus. We go in as the balloon animal duo. I’ll twist a poodle. You distract the parents with small talk.” “You want me to make small talk with people who haven’t slept in three years? What’s the topic? The existential dread of a soggy diaper?” “You’re overthinking. You just say ‘cute kid’ and point at their toddler. Works every time.” “Fine. But when we escape, I’m not carrying the giant teddy bear. It’s got dead eyes, Bill. It’s seen things.” “The bear is worth three hundred on eBay. You’ll carry it, and you’ll smile.” “And if the birthday boy catches me?” “You hide behind me and whisper, ‘Hey, wanna see a magic trick where a present disappears?’” “That’s terrible. That’s a felony with a side of therapy.” “Welcome to the big leagues, Molly. Now pass me the glitter glue. I need a poodle with pizazz.”
created_at
"2026-04-25T14:38:39.054Z"
4/25/2026, 2:38:39 PM
model
"deepseek/deepseek-v4-flash-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
723
moderation_latency
(null)
generation_time
4614
tokens_prompt
77
tokens_completion
329
native_tokens_prompt
67
native_tokens_completion
296
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
0.00009226
router
(null)
provider_responses
0
endpoint_id
"934a69f9-bd54-474b-beca-24560f721e12"
is_byok
false
latency
55
model_permaslug
"deepseek/deepseek-v4-flash-20260423"
provider_name
"DeepInfra"
status
(null)
1
endpoint_id
"722e0746-2034-4d6e-a08f-e6271849ee9e"
id
"8520f0c7-d5a6-4b27-a5ad-e53992bf696c"
is_byok
false
latency
278
model_permaslug
"deepseek/deepseek-v4-flash-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-1777127919-19Dbii2x1B7nEDwsEUlo"
session_id
(null)
api_type
"completions"
id
"gen-1777127919-E3gyPet2Pd3XyzWEzGzG"
upstream_id
"8520f0c7-d5a6-4b27-a5ad-e53992bf696c"
total_cost
0.00009226
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
99.38%
Matches word count
n/a
neededClean
false
words
195
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
197
wordsDialogue
197
68.0824%