NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5ZFR03MA84XAA65399FR7
Run Details
Model:
DeepSeek V3 (2024-12-26)
Test:
Dialogue tags
Scenario:
Write unattributed dialogue
Duration:
10.2s
Cost:
$0.000302
Provider:
DeepInfra
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 softly as the morning sunlight spilled across the worn wooden floor. The scent of cinnamon and freshly baked bread hung in the air. “Did you hear about the commotion last night? Over at the library?” “Oh, I heard bits and pieces. Something about a broken window?” “Not just broken—shattered! And a book missing. Oddest thing. Who steals a book in this day and age?” “Which book?” “That old leather-bound copy of ghost stories. You know, the one people say is cursed.” “Cursed? Really? That sounds like Mrs. Harper’s doing. She loves a good tale.” “Maybe, but the librarian swears she saw footprints leading away from the window. Small ones, like a child’s.” “A child? At midnight? That’s odd. Did they check the security cameras?” “Cameras were out. Storm knocked them out the day before. Convenient, isn’t it?” “Too convenient. You think someone’s playing a prank?” “Or trying to stir up trouble. You know how folks around here love a mystery.” “Well, if it is a prank, they’ve got the whole town talking. Honestly, though, I’m more concerned about whether there’s any of that blueberry crumble left.” “You’re in luck. I saw two slices behind the counter. Shall we split one?” “Absolutely. Mysteries can wait. Blueberry crumble cannot.”
created_at
"2026-02-16T21:34:01.982525+00:00"
2/16/2026, 9:34:01 PM
model
"deepseek/deepseek-chat-v3"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
295
moderation_latency
(null)
generation_time
10115
tokens_prompt
81
tokens_completion
372
native_tokens_prompt
64
native_tokens_completion
316
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)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.00030172
router
(null)
provider_responses
0
id
"chatcmpl-RH9E935isq1dGyAWfSU1ahlJ"
status
200
is_byok
false
latency
46
endpoint_id
"5294d55f-9012-496b-8f22-8cc919432dcd"
provider_name
"DeepInfra"
model_permaslug
"deepseek/deepseek-chat-v3"
api_type
"completions"
id
"gen-1771277631-d4aDVw0WtDjeU4swKdxz"
upstream_id
"chatcmpl-RH9E935isq1dGyAWfSU1ahlJ"
total_cost
0.00030172
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Count dialogue tags
n/a
neededClean
false
dialogueTags
(empty)