NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5E7DDG3Z1Q5R8BEZDTY6Q
Run Details
Model:
DeepSeek V3.2
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
11.2s
Cost:
$0.000138
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
The planning table was a sprawl of sugar-fueled chaos. A half-eaten cupcake served as a paperweight for the schematics of a bouncy castle. Molly, peering through binoculars crafted from toilet paper tubes, tracked the patrol pattern of a determined five-year-old with a security badge made of crayon. “Perimeter’s soft,” she muttered, tapping a map drawn in glitter glue. “But the main objective is a fortress. It’s in the living room, under the twinkle lights.” Bill, adjusting a fake mustache that kept coming unstuck, squinted at the diagram. “The Cake Guardian is the real variable. She’s got eyes in the back of her head and a lethal spatula.” Molly nodded, her expression grave. “We go in during the piñata frenzy. Distraction is key. You create a juice-box spill by the slide. I’ll belly-crawl through the ball pit.” “What about the attack dog?” “The shih tzu? His name is Mr. Wiggles. A well-placed chew toy will neutralize him.” She pointed a finger, her voice dropping to a whisper. “Remember, Bill, no fingerprints. We wear the silly putty gloves. This is a clean operation.” Bill sighed, the weight of the mission upon him. “I just hope the big one is the racing car set and not another educational puzzle.” “Focus,” Molly hissed. “The birthday boy is blowing out the candles in ten. Places.”
created_at
"2026-02-16T21:24:37.720967+00:00"
2/16/2026, 9:24:37 PM
model
"deepseek/deepseek-v3.2-20251201"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3016
moderation_latency
(null)
generation_time
11119
tokens_prompt
77
tokens_completion
357
native_tokens_prompt
73
native_tokens_completion
313
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
0.00013792
router
(null)
provider_responses
0
status
429
is_byok
false
latency
55
endpoint_id
"f270e10b-1d21-4345-9e4e-662ddb5ba441"
provider_name
"DeepInfra"
model_permaslug
"deepseek/deepseek-v3.2-20251201"
1
id
"7816a3eb625c4b2999eb610dadc2a5aa"
status
200
is_byok
false
latency
3016
endpoint_id
"3570a950-b80f-4acd-bb92-8e1c1c51fafc"
provider_name
"AtlasCloud"
model_permaslug
"deepseek/deepseek-v3.2-20251201"
api_type
"completions"
id
"gen-1771277065-fEf1r1FofgTEewVUIRiQ"
upstream_id
"7816a3eb625c4b2999eb610dadc2a5aa"
total_cost
0.00013792
cache_discount
(null)
upstream_inference_cost
0
provider_name
"AtlasCloud"
Evaluation details
Result
Evaluator
Details
Meta Data
0.4920%
Matches word count
n/a
neededClean
false
words
227
0.0000%
Dialogue to Total Word Ratio
Ratio: 56.39%, Deviation: 46.39%
neededClean
false
wordsTotal
227
wordsDialogue
128
0.2460%