NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2G3SJMYP70H873VMCTAVJ8
Run Details
Model:
DeepSeek V4 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
5.7s
Cost:
$0.000118
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
95.99%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
66.3873%
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, focus. The target is a ten-year-old’s birthday party, not Fort Knox.” “I’m focused, Molly. I just don’t think my knees can handle another bouncy castle infiltration.” “Your knees aren’t the point. The presents are stacked right by the cake table. We slip in during the ‘Happy Birthday’ song.” “And if a kid catches us?” “We smile, wave, and say we’re the magician’s assistants. You’ve got the glitter vest, I’ve got the rabbit.” “That rabbit bit me last time.” “He’s been sedated. Look, the window to the playroom is unlocked. We climb, we grab, we vanish.” “What about the parents? They’ll be watching like hawks over the goodie bags.” “The parents will be drunk on fruit punch and regret. Trust me. I’ve studied their patterns.” “You studied parents at a kid’s party?” “I’m a professional, Bill. Now, when I give the signal—a triple honk of my novelty party horn—you go for the big box with the bow.” “The one that says ‘To Timmy’?” “That one. It’s probably a tablet. Or a drone.” “Or a stuffed llama.” “Llama, tablet, doesn’t matter. It’s profit. Now, repeat the plan.” “Sneak in, avoid children, grab presents, run.” “And?” “And don’t get hit with glitter.” “Attaboy. Let’s go steal some innocence.”
created_at
"2026-04-25T14:18:49.305Z"
4/25/2026, 2:18:49 PM
model
"deepseek/deepseek-v4-flash-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
552
moderation_latency
(null)
generation_time
5585
tokens_prompt
77
tokens_completion
406
native_tokens_prompt
67
native_tokens_completion
388
native_tokens_completion_images
(null)
native_tokens_reasoning
50
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
0.00011802
router
(null)
provider_responses
0
endpoint_id
"722e0746-2034-4d6e-a08f-e6271849ee9e"
id
"1faa5d9f-f12b-41f5-bb9d-0403b4924b7c"
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-1777126729-QGPMdvcFsg1Wt9BRzDtL"
session_id
(null)
api_type
"completions"
id
"gen-1777126729-Kzh7T9reR2BWix6tH2Z9"
upstream_id
"1faa5d9f-f12b-41f5-bb9d-0403b4924b7c"
total_cost
0.00011802
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
95.99%
Matches word count
n/a
neededClean
false
words
208
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
210
wordsDialogue
210
66.3873%