NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2DJ9E5PZK0FRBDHK4WXBB8
Run Details
Model:
DeepSeek V4 Pro
Test:
Dialogue tags
Scenario:
Write unattributed dialogue
Duration:
11.5s
Cost:
$0.000338
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 warm, yeasty air of Rise & Shine Bakery wrapped around me like floured arms. Outside the picture window, Maple Street glittered with a fresh dusting of snow. Inside, the display case glowed, a jewel box of glossy fruit tarts and pillowy cinnamon rolls. “You’ve been staring at that same scone for five minutes. Did it confess to the crime?” “It’s a cranberry-orange scone, not a witness. I’m just… thinking. The texture is all wrong for a proper bludgeon. Too crumbly. No structural integrity.” “Only you would analyze the murder weapon potential of a baked good. The police are saying it was the brass candlestick, same one from the library window. Case closed.” “Is it? Then why was there a dusting of flour on the rug near the body? Not just any flour. A fine-milled pastry flour. The kind Millie uses for her scones.” “Millie wouldn’t hurt a fly and you know it. She was here, glazing windows at five in the morning.” “With an apron full of this exact flour, which, I might add, is completely inconsistent with the crime scene’s candlestick narrative. The angle of the wound was all wrong for a swing. It was a thrust. Almost certainly a secondary object, something heavy but with a smaller, concentrated impact point. Like a frozen stick of butter. Or a rolling pin.” “You’re telling me a frozen stick of butter is the real murder weapon?” “I’m telling you that candlestick was a plant. And whoever put it there doesn't know a crumb coat from a crumb trail.”
created_at
"2026-04-25T13:34:18.569Z"
4/25/2026, 1:34:18 PM
model
"deepseek/deepseek-v4-pro-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
896
moderation_latency
(null)
generation_time
11464
tokens_prompt
81
tokens_completion
399
native_tokens_prompt
65
native_tokens_completion
356
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
0.000337995
router
(null)
provider_responses
0
endpoint_id
"d084170b-dd99-4e53-9ae1-b63881f3cfbf"
id
"981b3329-b5d2-4b3a-9710-eb3b8e0cc9df"
is_byok
false
latency
299
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-1777124058-0xx4YHTcBfOvc28gfPu8"
session_id
(null)
api_type
"completions"
id
"gen-1777124058-oUYAimbDQahdNtlAEhR2"
upstream_id
"981b3329-b5d2-4b3a-9710-eb3b8e0cc9df"
total_cost
0.000337995
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)