NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-10-01M25DXXQ5FPGT3K2KEAXSRCR4
Run Details
Model:
DeepSeek V4.1 Flash (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
44.0s
Cost:
$0.006766
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
98.71%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
99.3552%
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 van’s dashboard like a treasure map. The glittery invitation promised cake and chaos. Bill leaned over, smelling faintly of stolen frosting. Outside the community center, a bouncy castle wheezed while children screamed in sugar-fed joy. A clown in a too-tight wig honked a bicycle horn and promised balloon poodles. Somewhere inside, a cake was waiting. “This is the score,” Molly said. “It’s a birthday party,” Bill said. “Exactly. No armed guards. No vault. Just thirty kids, six exhausted parents, and a gift table nobody watches.” Bill rubbed his chin. “There’s a clown.” “He’s with us.” “You hired a clown?” “I hired a clown who owes me money. He’ll do a magic trick, make a rabbit disappear, and then make the parents disappear into the cake line.” Bill looked unconvinced. “What about the presents?” Molly unfolded a crayon sketch of the community center. She tapped the kitchen door, the back hallway, and the folding table stacked with wrapped boxes. “I go in as a caterer. You go in as my assistant. We grab the gifts, load them into the van, and leave before the birthday song.” “Why am I always the assistant?” “Because you ate the evidence.” “It was a cupcake.” “It was the grandmother’s cupcake.” Bill sighed. He remembered the grandmother, a tiny woman with a cane and a stare that could strip paint. She had accused the clown of theft. The clown had accused a pigeon. The pigeon had flown away with a frosting rose. It had been a messy reconnaissance. Molly checked her watch. The party had entered its chaotic middle phase. Pin the tail on the donkey had ended in a shoving match. Musical chairs had produced two tears and one bruised shin. The parents had retreated to the punch bowl, where they discussed school districts and gluten-free snacks. Only one father remained on guard, but he was asleep in a folding chair with a paper plate on his chest. “What if a kid sees us?” Bill asked. “Kids are short,” Molly said. “We’ll tell them we’re present inspectors.” “They’ll believe that?” “Bill, yesterday you believed a pigeon was a drone.” “It had a camera.” “It had a cracker.” Bill pulled on his gloves. They were yellow rubber gloves, the kind used for dishes and, apparently, for stealing from children. Molly adjusted her catering cap and lifted a tray of empty cups. The plan was simple: walk in, look busy, grab the gifts, walk out. If anyone asked, they were with the caterer. If anyone asked twice, they were with the clown. If anyone asked three times, Bill was prepared to fake a stomach cramp. The clown began singing a song about a duck. Children shrieked. A piñata creaked overhead like a piñata. Molly opened the van door. Bill followed, carrying a box of napkins and a heavy conscience. They slipped toward the community center, ready to ruin a birthday and possibly a childhood.
created_at
"2026-09-10T10:31:03.145Z"
9/10/2026, 10:31:03 AM
model
"deepseek/deepseek-v4.1-flash-20260910"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
734
moderation_latency
(null)
generation_time
44010
tokens_prompt
75
tokens_completion
11126
native_tokens_prompt
93
native_tokens_completion
11253
native_tokens_completion_images
(null)
native_tokens_reasoning
10593
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
0.00676575
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1789036263-ydn1BXUoei2IQVQcAoSo"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1789036263-WQq223oArYtWyQncJUeH"
upstream_id
"9b2ff599-87ad-4496-82c8-ad54620a051a"
provider_responses
0
endpoint_id
"fd104bc8-535a-4d31-a847-a8e402e86a02"
id
"9b2ff599-87ad-4496-82c8-ad54620a051a"
is_byok
false
latency
336
model_permaslug
"deepseek/deepseek-v4.1-flash-20260910"
provider_name
"DeepSeek"
status
200
total_cost
0.00676575
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
98.71%
Matches word count
n/a
neededClean
false
words
494
100.00%
Dialogue to Total Word Ratio
Ratio: 29.98%, Deviation: 0.02%
neededClean
false
wordsTotal
497
wordsDialogue
149
99.3552%