NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2ECKYC6MB8KXGGTRE574NS
Run Details
Model:
DeepSeek V4 Flash
Test:
Dialogue tags
Scenario:
Write unattributed dialogue
Duration:
4.3s
Cost:
$0.000083
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 bell above the bakery door chimed, then fell silent. A chair scraped against the worn wood floor. “He used salted butter.” “I don’t bake with anything else.” “That’s not the point. The point is, for a butter tart to win the county fair, you use unsalted. It’s tradition. My grandmother’s recipe says so, right there in the margin.” A warm, yeasty silence settled between them. The espresso machine hissed softly from the corner. “Your grandmother also said a good liar always adds one extra clove to their pumpkin pie.” “That was about covering up a poor harvest, not about… this.” “Was it, now?” The sound of a teacup being set down, a gentle clink. “Or was she telling you that rules are just suggestions dressed up in a starched apron?” “It’s not a costume. It’s integrity.” “Integrity is knowing why you’re breaking the rule.” A pause, the rustle of a paper bag. “His tart won because it tasted better. Not because it was more correct. Taste wins, Margaret. Rules just argue.” The chair creaked again, then a soft thud of a door closing. The bell chimed once more.
created_at
"2026-04-25T13:48:41.297Z"
4/25/2026, 1:48:41 PM
model
"deepseek/deepseek-v4-flash-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
572
moderation_latency
(null)
generation_time
4160
tokens_prompt
81
tokens_completion
307
native_tokens_prompt
65
native_tokens_completion
263
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
0.00008274
router
(null)
provider_responses
0
endpoint_id
"722e0746-2034-4d6e-a08f-e6271849ee9e"
id
"01c60db8-5531-4d92-a436-d77fa0129063"
is_byok
false
latency
283
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-1777124921-opLWRpEmZMy6HuheYw2h"
session_id
(null)
api_type
"completions"
id
"gen-1777124921-F6MWySuFicHlu0sfWu0f"
upstream_id
"01c60db8-5531-4d92-a436-d77fa0129063"
total_cost
0.00008274
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)