NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KBEXK6FBYS2E057V99E735
Run Details
Model:
DeepSeek V4 Flash 0731 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
8.4s
Cost:
$0.000214
Provider:
Makora
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 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
The backyard was a battlefield of primary colors, a gaudy explosion of crepe paper streamers and a banner that screamed "HAPPY 7TH BIRTHDAY, CHLOE!" in glittering, nausea-inducing pink. Molly, squatting behind a hedge of prickly holly, peered through a pair of opera glasses that were both far too fancy and far too ridiculous for the task. Bill, crouched beside her, was trying to pick a piece of gum off the sole of his shoe without making a sound. “Okay, I count four adults,” Molly whispered, her voice a low, dramatic rasp. “One is neck-deep in a cooler of juice boxes, one is on the phone looking furious, and two are attempting to referee a tug-of-war over a bouncy castle. That leaves the present table—a veritable fortress of wrapped delight—completely unguarded.” “It’s not unguarded,” Bill muttered, finally freeing the gum and flicking it into a bush. “There’s a kid sitting under the table. Looks about three. He’s building a tower out of his own snot.” “That’s not a guard, Bill. That’s a biological hazard. And we have protection.” Molly tapped the rubber nose on her clown costume. “We are Bubbles and Chuckles, the world’s most beloved party entertainers. They invited us.” “They invited a clown,” Bill said, tugging at his polka-dotted collar. “Not a forty-year-old man with a suspicious limp and a deep-seated fear of balloons. What if I have to juggle?” “You won’t have to juggle,” Molly said, exasperated. “You have to set off the confetti cannon. The distraction. Then, I swoop in, arms wide, and the presents go into the magical clown sack. Simple.” Bill stared at the pile of gifts. “They’re wrapped in Frozen paper. I feel like a war criminal.” “Feel like a wealthy one,” Molly countered, her eyes gleaming. “Look at that box on top. It’s huge. It’s either a dollhouse or a small car. Either way, it’s our retirement fund.” “A child is now crying,” Bill observed. “Which one?” Molly asked, adjusting her wig. “The birthday girl. She’s pointing at us.” Molly froze. “She’s pointing at *us*? Or at the hedge?” “She’s pointing at you, and she’s saying ‘The clown lady has a beard.’” Bill sighed. “You forgot to shave your chin again.” “It’s not a beard, it’s a shadow. It adds character. Now, listen,” Molly said, grabbing Bill’s wrist. “Here’s the new plan. You go in first. Trip over your own feet. Fall into the cake. Create maximum chaos. While everyone is screaming about the smashed fondant, I’ll be under the table with the snot kid, doing the swap.” “Swap? We’re not swapping. We’re taking.” “We’re swapping the presents with these,” Molly said, pulling a handful of shiny, empty boxes from her bag. “Weights and looks. By the time they open them tonight, we’ll be on a beach in Aruba. The kid will just think his parents bought him a really nice brick.” Bill looked at the shiny box in his hand. “It’s not a brick. It’s a box of rocks I collected from the driveway.” “Precisely. The perfect crime.” Molly shoved him forward. “Go get ‘em, Chuckles.” “I hate you,” Bill said, but he stood up, adjusted his bow tie, and took a deep breath. “I know, Bubbles,” Molly whispered, grinning. “I know.” She watched him waddle toward the party, his limp suddenly becoming much more pronounced, a look of pure, terrified determination on his face. In the distance, the tiny girl named Chloe began to wail, and Molly felt a strange, sudden pang of… something. It passed quickly when she estimated the retail value of the Frozen-themed wrapping. She ducked back down, ready to make her move.
created_at
"2026-09-03T10:01:34.584Z"
9/3/2026, 10:01:34 AM
model
"deepseek/deepseek-v4-flash-20260731"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
153
moderation_latency
(null)
generation_time
7994
tokens_prompt
75
tokens_completion
1146
native_tokens_prompt
67
native_tokens_completion
1069
native_tokens_completion_images
(null)
native_tokens_reasoning
185
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
0.000214485
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788429694-7s9mULK89GDbf0baBfcJ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788429694-rTg1qn0k5Prgw1Oxfx6j"
upstream_id
"chatcmpl-769e40a5-80ca-43ae-a4de-b25d35f70614"
provider_responses
0
endpoint_id
"562d5188-97d1-49f7-8881-d0e417e31ada"
id
"chatcmpl-769e40a5-80ca-43ae-a4de-b25d35f70614"
is_byok
false
latency
110
model_permaslug
"deepseek/deepseek-v4-flash-20260731"
provider_name
"Makora"
status
200
total_cost
0.000214485
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Makora"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
607
0.00%
Dialogue to Total Word Ratio
Ratio: 59.09%, Deviation: 29.09%
neededClean
false
wordsTotal
616
wordsDialogue
364
0.0000%