NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2EHZGFXPKPRVYJXQFXC008
Run Details
Model:
DeepSeek V4 Pro
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
10.2s
Cost:
$0.000350
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
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
The air-conditioned roar of a bouncy castle was doing its level best to drown out their scheming. Molly lowered her binoculars, her gaze fixed on the inflatable purple dragon shuddering across the lawn. "So," Bill whispered, licking buttercream from his thumb, "the cake is a lie. The presents are the real target." "Obviously, Bill. No one puts a solid-gold, limited-edition Princess Puffle-Muffin tiara on a cupcake tower. The gift table is a magpie’s manifesto." "Right. I go in as Giggles the Clown. A honk for entry, a honk for distress." "You can’t juggle." "My emotional distress is the entertainment. You’re on cake duty." "Page three of the playbook: 'The Candle Catastrophe.' I trip, I wail, a $400 sheet cake becomes a Jackson Pollock. In the chaos, you bag the tiara." "Solid. But how do I neutralise the birthday girl? She’s a tiny, terrifying gatekeeper. Look at her, guarding that pile of loot like a dragon with a sugar high." Molly smiled, pulling a shimmering, detachable mermaid tail from her bag. "I’ve got a prop. I’ll tell her the cake explosion was a mermaid curse, and we need to go on a magical quest for a mop. By the time her existential crisis is over, we’ll be in a taxi." "A quest for a mop. That’s either genius or a fever dream." "It’s an exit strategy, Bill. Now, honk twice if you see a piñata bat. Those things are a wildcard."
created_at
"2026-04-25T13:51:36.979Z"
4/25/2026, 1:51:36 PM
model
"deepseek/deepseek-v4-pro-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1244
moderation_latency
(null)
generation_time
10117
tokens_prompt
77
tokens_completion
362
native_tokens_prompt
67
native_tokens_completion
369
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
0.000350175
router
(null)
provider_responses
0
endpoint_id
"d084170b-dd99-4e53-9ae1-b63881f3cfbf"
id
"b9c1af93-7e09-4aec-b024-91ba0f86f07f"
is_byok
false
latency
368
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-1777125096-SO07QnDImrpoerBhmYOJ"
session_id
(null)
api_type
"completions"
id
"gen-1777125096-lHePrz6XiZ2tgLNZpwM1"
upstream_id
"b9c1af93-7e09-4aec-b024-91ba0f86f07f"
total_cost
0.000350175
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
240
0.00%
Dialogue to Total Word Ratio
Ratio: 78.69%, Deviation: 28.69%
neededClean
false
wordsTotal
244
wordsDialogue
192
0.0000%