NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2HVEK95RX47MQ40GRQ4DW4
Run Details
Model:
DeepSeek V4 Flash
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
4.8s
Cost:
$0.000091
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
68.10%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
34.0511%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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
Molly scraped frosting from her nail with a toothpick. “The bounce house is a distraction, Bill. Not the target.” Bill adjusted his balaclava, which smelled faintly of catnip. “But what if I want to bounce?” “You’re forty-seven.” She unrolled a blueprint of the Dinosaur Safari Kingdom Party Zone. The toilet block was circled in red Sharpie. “The loot is in the Goody Bag Fortress. Past the balloon arch and the face-painting station. The clown, Bubbles, is a master of sleight-of-hand. He hides the presents in the inflatable T-Rex’s tail.” Bill squinted. “T-Rex has a tail?” “Model-Z. It’s a structural vulnerability.” She tapped the diagram. “You create a diversion by releasing the ladybug catapult’s spring. I slip in, grab the haul, exit through the ball pit. We rendezvous by the unicorn piñata.” A child shrieked with laughter outside. Bill chewed his lip. “But what if I get stuck in the ball pit again?” “Then I’ll leave you there.” Molly pocketed the blueprint. “That’s the plan.” “And Bubbles?” “He’ll be too busy crying about the catapult.” She pulled her ski mask down. “Let’s go make some five-year-olds miserable.”
created_at
"2026-04-25T14:49:13.069Z"
4/25/2026, 2:49:13 PM
model
"deepseek/deepseek-v4-flash-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
580
moderation_latency
(null)
generation_time
4744
tokens_prompt
77
tokens_completion
325
native_tokens_prompt
67
native_tokens_completion
293
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
0.00009142
router
(null)
provider_responses
0
endpoint_id
"722e0746-2034-4d6e-a08f-e6271849ee9e"
id
"50271722-d7c5-4643-b706-6ab4f86ce357"
is_byok
false
latency
343
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-1777128553-l6LzqU0SAIdRl8WgHhor"
session_id
(null)
api_type
"completions"
id
"gen-1777128553-bMXMaqTmxvkDDPMwoNWe"
upstream_id
"50271722-d7c5-4643-b706-6ab4f86ce357"
total_cost
0.00009142
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
68.10%
Matches word count
n/a
neededClean
false
words
186
0.00%
Dialogue to Total Word Ratio
Ratio: 67.19%, Deviation: 57.19%
neededClean
false
wordsTotal
192
wordsDialogue
129
34.0511%