NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2GNE4JQRQN4JPJH8T86HGT
Run Details
Model:
DeepSeek V4 Pro (Reasoning)
Test:
Dialogue tags
Scenario:
Write unattributed dialogue
Duration:
12.9s
Cost:
$0.000398
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
100.00%
Count dialogue tags
(details)
Model input
User/Human
You are an expert novel writer in the cozy mystery genre. Write a scene of about 200 words, with dialogue between two characters. The dialogue should be unattributed, meaning the speaker is not explicitly mentioned. The scene should be set in a small town bakery. Only return the story, nothing else.
Model response
Messages
Meta
AI
The scent of warm cinnamon and yeast hung thick in the air, a sweet curtain against the grey drizzle outside the window. Behind the counter, dust motes danced in the slant of light falling on a tray of cooling scones. “You can’t frost a lie, you know. The cracks will always show.” A wire whisk paused its lazy spin in a bowl of vanilla glaze. “It’s not a lie. It’s a careful arrangement of selected truths. There’s a difference.” “Is there? Because from where I’m standing, the careful arrangement you told the sheriff about finding poor Mr. Henderson looks an awful lot like a double batch of deception.” A soft, humorless laugh. “I simply said I found him when I came in to start the sourdough.” “At four in the morning. But Gladys Pumpernickel swears she saw a light on in the back kitchen just after midnight. She was walking her cat, you see. Insomniac tabby. She’s very reliable.” The whisk began to move again, a little too briskly, splashing a drop of white glaze onto the flour-dusted counter. “Gladys has an overactive imagination and a cat with a thyroid problem. My timeline is perfectly clear.” “Was it, then? Clear enough to explain why his pocket watch, the one he was never without, is currently tucked behind the jar of candied violets on the top shelf?”
created_at
"2026-04-25T14:28:27.414Z"
4/25/2026, 2:28:27 PM
model
"deepseek/deepseek-v4-pro-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
597
moderation_latency
(null)
generation_time
12852
tokens_prompt
81
tokens_completion
488
native_tokens_prompt
65
native_tokens_completion
425
native_tokens_completion_images
(null)
native_tokens_reasoning
122
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
0.000398025
router
(null)
provider_responses
0
endpoint_id
"d084170b-dd99-4e53-9ae1-b63881f3cfbf"
id
"2a1ee4fe-f1b0-459e-af79-1492683d37bd"
is_byok
false
latency
276
model_permaslug
"deepseek/deepseek-v4-pro-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-1777127307-LNkB8GYbu1bC6XLXDZ5l"
session_id
(null)
api_type
"completions"
id
"gen-1777127307-rdS5bkiWtXc8gABGwvrm"
upstream_id
"2a1ee4fe-f1b0-459e-af79-1492683d37bd"
total_cost
0.000398025
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Count dialogue tags
n/a
neededClean
false
dialogueTags
(empty)