NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2GCGKEDX484N4NMP7S6C8Q
Run Details
Model:
DeepSeek V4 Flash
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
5.4s
Cost:
$0.000098
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
99.92%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
68.3535%
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 absolute lunatic, we’re supposed to be professionals.” “We are professionals! When’s the last time a bouncy castle was a security detail, huh? It’s genius.” “It’s a ten-foot castle shaped like a pirate ship. I can’t scale that in heels.” “Don’t wear heels. Wear clown shoes.” “I will not wear clown shoes.” “Fine, wear pumps, but you’ll miss the psychological warfare. I’ve got a horn that farts.” “A what? No. No horns. No farting. We case the joint, snatch the haul, and ghost.” “See, that’s your problem, Molly. You don’t appreciate the art of the distraction. I’ll do the balloon sword thing, you slip the Xbox behind the magician’s rabbit cage, we split the loot.” “The loot is a pile of unwrapped Play-Doh and a unicorn nightlight.” “Don’t underestimate the unicorn nightlight, Moll. That thing’s vintage. I saw one go for sixty on eBay.” “Sixty? Fine. But if we get caught by a six-year-old named Kevin, I’m telling them you took the last juice box.” “Kevin’s the one with the laser eyes. I’ll handle him. You just don’t trip over the face painter.” “I hate you.” “You love me. Now, let’s go steal a pony piñata.”
created_at
"2026-04-25T14:23:35.029Z"
4/25/2026, 2:23:35 PM
model
"deepseek/deepseek-v4-flash-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
731
moderation_latency
(null)
generation_time
5295
tokens_prompt
77
tokens_completion
342
native_tokens_prompt
67
native_tokens_completion
315
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
0.00009758
router
(null)
provider_responses
0
endpoint_id
"722e0746-2034-4d6e-a08f-e6271849ee9e"
id
"ae36995a-8ccd-4802-b7df-3905224bc33e"
is_byok
false
latency
387
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-1777127015-tHWY2YOuc5doVGn011ML"
session_id
(null)
api_type
"completions"
id
"gen-1777127015-WYdWv1867G1CDNgPISNR"
upstream_id
"ae36995a-8ccd-4802-b7df-3905224bc33e"
total_cost
0.00009758
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
99.92%
Matches word count
n/a
neededClean
false
words
197
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
201
wordsDialogue
201
68.3535%