NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-10-01M25DVX5FGHN0KXJMXPR0DNK3
Run Details
Model:
DeepSeek V4.1 Flash (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
54.9s
Cost:
$0.009618
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
43.38%
Matches word count
(details)
99.02%
Dialogue to Total Word Ratio
(details)
71.2006%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 spread the birthday invitation across the sticky Waffle Hut table, using a butter knife to pin down one corner. The card showed a cartoon superhero, a rainbow, and the words Twins Turn Seven. Bill leaned over it, wearing a party hat he had stolen from a wake and eating a cupcake he had stolen from a baby shower. Outside, rain smeared the window. Inside, their plan smelled like maple syrup and bad decisions. "The Rodriguez twins," Molly said. "Two o'clock. Forty kids, two bouncy castles, one magician, and a pile of presents in the dining room." "Bouncy castles are a nightmare," Bill said. "Static. Screaming. Last time I bounced, I lost a shoe and a tooth." Molly ignored him and tapped the invitation. The magician would perform at two fifteen, pulling a rabbit from a hat and then making it vanish. While every child watched the rabbit not exist, Molly and Bill would back a van up to the side gate. Bill would carry sacks. Molly would smile. They would load the presents and drive away before the cake was cut. "What kind of loot?" Bill asked. "Stuffed dinosaurs? Scented markers?" "One present is a limited-edition robot dog," Molly said. "Resells for eight hundred. Another is a gold bracelet from Dubai. The rest goes to a guy who runs a daycare." The waitress refilled their coffee. Molly waited until she left. They needed disguises. The clown already had two assistants, teenagers named Kayla and Brayden. Molly planned to offer them fifty bucks and a vape to disappear. Bill would wear a rainbow wig. Molly would carry a clipboard and pretend to be the party planner. Their disguises were simple: a clipboard for Molly, a rainbow wig for Bill, and the confidence of people who had never been asked to leave a party. The van would wait by the dumpsters. They rehearsed their smiles. If anyone asked questions, Bill would say, "Wow, that's a big present," and "Please don't tell your parents." "What about security?" Bill asked. "A face painter with a taser. Ex-military aunt." Bill put down his cupcake. "We'll be in costume. They'll assume we belong." "And if the clown recognizes us?" "He won't. He's legally blind in one eye and emotionally blind in the other." Bill considered this. He had robbed banks, casinos, and one extremely defensive hamster, but never a children's party. The stakes felt stickier. Still, the robot dog would pay his rent. The bracelet would pay Molly's dental bills. The rest would buy them a quiet week in a motel with cable. "You carry the sacks," Molly said. "You smile. You say 'Wow, that's a big present' and 'Please don't tell your parents.'" Bill looked at the invitation. "What if the robot dog is already opened?" "Then we steal the batteries." Bill stood and adjusted his party hat. For a criminal mastermind, Molly had a strange moral code. She called it branding. They left the Waffle Hut. Behind them, the waitress picked up the invitation and called her sister, whose twins were turning seven. The party was about to get interesting.
created_at
"2026-09-10T10:29:57.044Z"
9/10/2026, 10:29:57 AM
model
"deepseek/deepseek-v4.1-flash-20260910"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
523
moderation_latency
(null)
generation_time
54855
tokens_prompt
75
tokens_completion
14948
native_tokens_prompt
93
native_tokens_completion
16007
native_tokens_completion_images
(null)
native_tokens_reasoning
15328
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"
service_tier
(null)
usage
0.00961815
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1789036197-uHUjsdXt9qTbv15xy1c1"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1789036197-xI4aFlipYRhzrIWV13Oj"
upstream_id
"395d1bc8-8208-4430-afcc-ba10e4f72f5d"
provider_responses
0
endpoint_id
"fd104bc8-535a-4d31-a847-a8e402e86a02"
id
"395d1bc8-8208-4430-afcc-ba10e4f72f5d"
is_byok
false
latency
347
model_permaslug
"deepseek/deepseek-v4.1-flash-20260910"
provider_name
"DeepSeek"
status
200
total_cost
0.00961815
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
43.38%
Matches word count
n/a
neededClean
false
words
517
99.02%
Dialogue to Total Word Ratio
Ratio: 30.44%, Deviation: 0.44%
neededClean
false
wordsTotal
519
wordsDialogue
158
71.2006%