NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5GY5THD4FPNN41M1J7Z86
Run Details
Model:
DeepSeek V3.1
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
9.1s
Cost:
$0.000336
Provider:
AtlasCloud
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 stale air of the van was thick with the scent of old coffee and desperation. Molly tapped a crayon-drawn map of the McAllister residence with grim finality. "Target is a seven-year-old named Timmy. The haul is a mountain of brightly coloured boxes currently residing under a 'Happy Birthday' banner in the den." Bill, peering through binoculars at the house, lowered them slowly. "Right. Den. Okay. But Molly, the perimeter is swarming with tiny, sugar-fueled hostiles. And a clown. A very sad-looking clown." "He's the diversion," Molly insisted, circling the clown's position on the lawn. "His name is Bubbles. You create a minor disturbance with the balloon animals—pretend to have a moral crisis—and I slip in through the patio doors." "A moral crisis? Molly, he's a clown. He probably has one every Tuesday." Bill sighed Bill. "And what's the exit strategy?" Molly smiled, holding up a small, foil-wrapped object. "The piñata. We grab the loot, whack the donkey, and escape in the ensuing candy-fueled chaos." Bill stared at the candy-filled donkey, then back at the house where a group of children were now vigorously assaulting it with a plastic bat. "You know," he said, "for a den of thieves, we've hit a new low." "It's not a low, Bill," Molly said, pulling a black beanie over her head. "It's a strategic repositioning in an undervalued market. Now, go upset the clown."
created_at
"2026-02-16T21:26:04.288759+00:00"
2/16/2026, 9:26:04 PM
model
"deepseek/deepseek-chat-v3.1"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
885
moderation_latency
(null)
generation_time
9054
tokens_prompt
95
tokens_completion
359
native_tokens_prompt
68
native_tokens_completion
332
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)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.0003358
router
(null)
provider_responses
0
id
"d3134239b26349b686450aa37bef9ae7"
status
200
is_byok
false
latency
885
endpoint_id
"0bfe3535-ed02-47ea-a98f-60cc8af4fab2"
provider_name
"AtlasCloud"
model_permaslug
"deepseek/deepseek-chat-v3.1"
api_type
"completions"
id
"gen-1771277154-L3Aru5LgoQnpumILh7QK"
upstream_id
"d3134239b26349b686450aa37bef9ae7"
total_cost
0.0003358
cache_discount
(null)
upstream_inference_cost
0
provider_name
"AtlasCloud"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0000%
Matches word count
n/a
neededClean
false
words
241
40.4917%
Dialogue to Total Word Ratio
Ratio: 59.75%, Deviation: 9.75%
neededClean
false
wordsTotal
241
wordsDialogue
144
20.2459%