NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2FMRDTBD2WTP5V5Q7QQR6Y
Run Details
Model:
DeepSeek V4 Flash
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
5.2s
Cost:
$0.000098
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
97.63%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
48.8138%
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 crouched behind a bouncy castle, squinting at the gift table. “This is insane,” she whispered. “We’re stealing from seven-year-olds.” Bill adjusted his clown wig, checking his makeup in a compact mirror. “It’s not stealing from them, it’s re-distributing joy. Their parents spent too much. We’re restoring balance.” “Balance? We’re wearing face paint and carrying a sack labeled ‘Swag.’” “It’s thematic,” he said, stuffing a rubber chicken into his pocket. “The birthday boy, Timmy, just opened a drone. You want that in the hands of a child who can’t even spell ‘drone’? I can sell that to a guy who owns a falcon.” “A falconer?” “Precisely. Now, your distraction. You see the piñata?” “The unicorn?” “The unicorn’s about to reveal its tragic, candy-less destiny. You’ll scream, ‘There’s a real spider!’ They’ll panic. I’ll execute the grab.” “That’s cruel.” “So is the economy,” Bill said, blowing a squeaky horn. “Ready?” Molly sighed, then waved. “Fine. But if we get arrested by a parent in a princess costume, I’m telling them you named the operation ‘Loot and Booty.’” “That was your idea,” Bill muttered, as a child tugged his sleeve. “Hold that thought, future fence.”
created_at
"2026-04-25T14:10:36.606Z"
4/25/2026, 2:10:36 PM
model
"deepseek/deepseek-v4-flash-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
605
moderation_latency
(null)
generation_time
5099
tokens_prompt
77
tokens_completion
357
native_tokens_prompt
67
native_tokens_completion
317
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
0.00009814
router
(null)
provider_responses
0
endpoint_id
"722e0746-2034-4d6e-a08f-e6271849ee9e"
id
"30f997c0-24ec-4f1e-9c71-65322a8f8346"
is_byok
false
latency
274
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-1777126236-EYPMbJsOoDRPCMFBS6Yz"
session_id
(null)
api_type
"completions"
id
"gen-1777126236-lEgcbhs6SlrBmiyowNE8"
upstream_id
"30f997c0-24ec-4f1e-9c71-65322a8f8346"
total_cost
0.00009814
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
97.63%
Matches word count
n/a
neededClean
false
words
193
0.00%
Dialogue to Total Word Ratio
Ratio: 73.33%, Deviation: 23.33%
neededClean
false
wordsTotal
195
wordsDialogue
143
48.8138%