NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5BJ41ZWJA0KXRAEM7C2P5
Run Details
Model:
DeepSeek V3.2
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
9.8s
Cost:
$0.000124
Provider:
AtlasCloud
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 squinted at the schematic of the suburban bungalow. “Bouncy castle complicates egress,” she muttered. Bill, polishing his fake clown nose, disagreed. “Nonsense. It’s perfect cover. I’ll be ‘Bubbles the Melancholy Clown,’ providing a distraction of profound sadness. The parents will be too busy offering me therapy to notice you.” He gestured to the party timeline. “Pinata strike is at 1500 hours. The sugar crash will induce mass hysterical fatigue in the juvenile units approximately twenty minutes later. That’s your window.” Molly studied the photograph of the loot: a pile of brightly wrapped boxes in the dining room. “Security?” “One neurotic golden retriever named Muffins. I’ve procured a decoy turkey.” He held up a suspiciously meat-like object. “And the real threat: a three-year-old ‘ninja’ named Timmy with a devastatingly accurate water pistol.” “So,” Molly said, tapping the blueprint. “You cause a somber spectacle, I neutralize Muffins with poultry, grab the goods, and we exfiltrate via the inflatable slide.” “A flawless plan.” Bill sighed, a tear of artistic commitment already welling in his eye. “They’ll never see the clown coming. Or going.”
created_at
"2026-02-16T21:23:08.714384+00:00"
2/16/2026, 9:23:08 PM
model
"deepseek/deepseek-v3.2-20251201"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1465
moderation_latency
(null)
generation_time
9804
tokens_prompt
77
tokens_completion
326
native_tokens_prompt
67
native_tokens_completion
281
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
0.0001242
router
(null)
provider_responses
0
id
"49b56b983bd44e76ad18920c0e6a0337"
status
200
is_byok
false
latency
1465
endpoint_id
"3570a950-b80f-4acd-bb92-8e1c1c51fafc"
provider_name
"AtlasCloud"
model_permaslug
"deepseek/deepseek-v3.2-20251201"
api_type
"completions"
id
"gen-1771276978-ELgjNfcbnMy4aIQMqChs"
upstream_id
"49b56b983bd44e76ad18920c0e6a0337"
total_cost
0.0001242
cache_discount
(null)
upstream_inference_cost
0
provider_name
"AtlasCloud"
Evaluation details
Result
Evaluator
Details
Meta Data
43.3783%
Matches word count
n/a
neededClean
false
words
183
0.0000%
Dialogue to Total Word Ratio
Ratio: 64.48%, Deviation: 54.48%
neededClean
false
wordsTotal
183
wordsDialogue
118
21.6892%