NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5S9SBFD02X4ZNEVKJ6B8V
Run Details
Model:
DeepSeek V3.2
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
9.7s
Cost:
$0.000141
Provider:
AtlasCloud
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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
“Okay, walk me through this one more time,” Bill whispered, peering through the binoculars at the brightly decorated house. “Simple,” Molly said, her eyes gleaming. “The clown arrives at three. We intercept him in the alley.” “We’re mugging a clown? For his giant shoes?” “For his giant shoes *and* his van. Then you, dressed as Clown #2, drive the van to the party. You’ll be let right in.” “And you?” “I’ll already be inside, disguised as the overly-enthusiastic aunt. I’ll create a diversion with the piñata. While the sugar-crazed mob is occupied, you load the presents from the gift table into the clown van.” “What’s the diversion?” “I’m going to ‘accidentally’ whack the birthday boy’s father with the bat. It’ll be chaos.” “And our exit?” “You drive. I’ll slip out the back, meet you at the rendezvous. We trade the clown van for our car and vanish.” Bill lowered the binoculars. “A clown van filled with stolen toys. This is a new low, Molly.” She patted his cheek. “It’s a high-volume, low-security score. Think of it as Santa’s evil twin.” “What if someone notices the real clown tied up in the alley?” “By then we’ll be halfway to the pawn shop. Now, can you honk a red nose? It’s key to the character.”
created_at
"2026-02-16T21:30:38.650325+00:00"
2/16/2026, 9:30:38 PM
model
"deepseek/deepseek-v3.2-20251201"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1237
moderation_latency
(null)
generation_time
9660
tokens_prompt
77
tokens_completion
361
native_tokens_prompt
73
native_tokens_completion
322
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
0.00014134
router
(null)
provider_responses
0
id
"654bf2eab7ac40519280825ea9e81d11"
status
200
is_byok
false
latency
1237
endpoint_id
"3570a950-b80f-4acd-bb92-8e1c1c51fafc"
provider_name
"AtlasCloud"
model_permaslug
"deepseek/deepseek-v3.2-20251201"
api_type
"completions"
id
"gen-1771277428-aSEIJYfAzLnLJLD3BJeF"
upstream_id
"654bf2eab7ac40519280825ea9e81d11"
total_cost
0.00014134
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
217
99.9874%
Dialogue to Total Word Ratio
Ratio: 88.94%, Deviation: 1.06%
neededClean
false
wordsTotal
217
wordsDialogue
193
71.6829%