NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2HPZT9NFTM24TQASHF99AX
Run Details
Model:
DeepSeek V4 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
6.2s
Cost:
$0.000118
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0014%
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
“Right, so the bouncy castle is our ingress,” Molly whispered, her voice a low hum over the pile of illustrated invitation cards on the coffee table. “The birthday boy, Timmy, turns six. The party is in the community hall. Security? A woman named Denise who wears a velour tracksuit and cries at cake-cutting.” Bill squinted at her. “We’re stealing presents from a six-year-old? That’s a new low, Mol. Even for us.” “No, it’s a new high. Low is stealing from nuns. This is a soft target. Denise loves unicorns and is allergic to confrontation. All we need is a distraction.” “What kind of distraction?” “I thought you’d like that part. You’re going to get stuck in the bouncy castle.” Bill’s jaw dropped. “Me?” “Yes. You hyperventilate, they deflate the castle to rescue you, chaos ensues, I bag the gifts. It’s a classic.” “I hate this plan.” “You hated the cat-burglar plan too. And the sewer diamond plan. And you were wrong about all of them. This is foolproof.” “The last foolproof plan landed me in a clown car with a raccoon.” “That raccoon was my backup. He did great. Now, focus. Are you in or out?” Bill sighed, rubbing his temples. “I’m in. But you’re buying me a drink after this, and not the cheap stuff.” “Deal. Now help me pick a spot to tie the fake unicorn to the ceiling fan.”
created_at
"2026-04-25T14:46:46.862Z"
4/25/2026, 2:46:46 PM
model
"deepseek/deepseek-v4-flash-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
547
moderation_latency
(null)
generation_time
6088
tokens_prompt
77
tokens_completion
428
native_tokens_prompt
67
native_tokens_completion
389
native_tokens_completion_images
(null)
native_tokens_reasoning
48
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
0.0001183
router
(null)
provider_responses
0
endpoint_id
"722e0746-2034-4d6e-a08f-e6271849ee9e"
id
"75517dd1-b202-4044-9a11-f2e7fd0b9c43"
is_byok
false
latency
279
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-1777128406-xQh0nw7QLVjoP1pAzOfF"
session_id
(null)
api_type
"completions"
id
"gen-1777128406-oXxYenRQVeuHk0FDbhsk"
upstream_id
"75517dd1-b202-4044-9a11-f2e7fd0b9c43"
total_cost
0.0001183
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
232
0.00%
Dialogue to Total Word Ratio
Ratio: 87.29%, Deviation: 37.29%
neededClean
false
wordsTotal
236
wordsDialogue
206
0.0014%