NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5DTBRTCYMVERGAG353NFD
Run Details
Model:
DeepSeek V3.1
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
8.2s
Cost:
$0.000284
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 traced a chubby crayon line on the crinkled map. "The objective is the Vandenberg living room," she whispered, her voice hushed with a gravity usually reserved for international jewel exchanges. "Codenamed: 'The Bouncy Castle Fortress'." Bill, chewing nervously on a plastic cocktail sword, nodded. "Intel confirms the primary security system is a Yorkshire Terrier named Mr. Snuffles. Highly volatile." The planning table was a masterpiece of low-budget espionage. Party invitations were surveillance photos, a half-eaten cupcake marked the kitchen exit, and a tower of LEGOs represented the gift pile. Molly pointed a stern finger at a diagram of a piñata. "The donkey is a diversion. Our window is during the sugar-fueled chaos post-destruction." She slid a tiny, glitter-filled backpack across the table. "Your gear. Containment for the loot." Bill peeked inside. "It's covered in unicorns." "Operational necessity," Molly stated, adjusting her landlady-disguise glasses. "We infiltrate as friendly aunts. You handle the distraction. I’ll secure the assets." Bill paled. "What's the distraction?" "You'll be leading the chicken dance." He swallowed hard. "The things we do for a box of Lego."
created_at
"2026-02-16T21:24:21.060886+00:00"
2/16/2026, 9:24:21 PM
model
"deepseek/deepseek-chat-v3.1"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
737
moderation_latency
(null)
generation_time
8166
tokens_prompt
95
tokens_completion
316
native_tokens_prompt
68
native_tokens_completion
277
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
0.00028355
router
(null)
provider_responses
0
id
"c5856a7d1f5246f39c45fe7eafc30858"
status
200
is_byok
false
latency
737
endpoint_id
"0bfe3535-ed02-47ea-a98f-60cc8af4fab2"
provider_name
"AtlasCloud"
model_permaslug
"deepseek/deepseek-chat-v3.1"
api_type
"completions"
id
"gen-1771277052-EwhEufMNE9YpdH6H2Wun"
upstream_id
"c5856a7d1f5246f39c45fe7eafc30858"
total_cost
0.00028355
cache_discount
(null)
upstream_inference_cost
0
provider_name
"AtlasCloud"
Evaluation details
Result
Evaluator
Details
Meta Data
68.1022%
Matches word count
n/a
neededClean
false
words
186
0.0000%
Dialogue to Total Word Ratio
Ratio: 45.70%, Deviation: 35.70%
neededClean
false
wordsTotal
186
wordsDialogue
85
34.0511%