NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2DQNNMRY4XDPQJQXEX8HCA
Run Details
Model:
DeepSeek V4 Flash
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
6.2s
Cost:
$0.000115
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
18.3940%
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’re telling me we need to case a bouncy castle?” “Molly, it’s not just a bouncy castle. It’s a fortress. Look at it. Inflatable walls, a moat of ball pit, and that slide? Tactical disadvantage.” “We’re stealing presents from a seven-year-old’s party. Not Fort Knox.” “Exactly. They let their guard down. Parents are distracted, sugar-crazed kids everywhere. Perfect cover. We blend in.” “Blend in? I’m five-foot-ten and covered in tattoos. Do you see any unicorns on me?” “I got us costumes. I’m going as a clown. You’re Princess Sparklehoof.” “I’m not wearing a tutu, Bill.” “It has a horn and a cape. And pockets. I had to sew them in myself. The loot goes in the rainbow bag.” “The loot. You mean a bunch of wrapped LEGO sets and a talking doll.” “The doll alone is worth forty bucks resale. Plus the gift cards. They always hide cash in cards, Molly.” “This is the dumbest plan you’ve ever had.” “And the easiest. We walk in, we smile, we hit the gift table while they’re doing the piñata. In, out, ten minutes.” “And if someone asks who I am?” “You tell them you brought the glow sticks. And then you hand them a glow stick. I bought two hundred.” “You bought glow sticks for a heist?” “The markup is incredible. And if the job goes south, we sell them on the corner. We’re diversified, Molly.” “We’re idiots.” “Profit-seeking idiots. Now put on the horn.”
created_at
"2026-04-25T13:37:14.937Z"
4/25/2026, 1:37:14 PM
model
"deepseek/deepseek-v4-flash-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
557
moderation_latency
(null)
generation_time
6040
tokens_prompt
77
tokens_completion
410
native_tokens_prompt
67
native_tokens_completion
376
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
0.00011466
router
(null)
provider_responses
0
endpoint_id
"722e0746-2034-4d6e-a08f-e6271849ee9e"
id
"64f2d231-16f9-4c43-98ee-5c181d833c10"
is_byok
false
latency
272
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-1777124234-ztq8xcH1itECESK7krL0"
session_id
(null)
api_type
"completions"
id
"gen-1777124234-iM4q6XvNv3K4cpQ3eT3b"
upstream_id
"64f2d231-16f9-4c43-98ee-5c181d833c10"
total_cost
0.00011466
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
244
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
249
wordsDialogue
249
18.3940%