NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5P9M89AK5A1HFAD1RC8VH
Run Details
Model:
DeepSeek V3.1
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
21.9s
Cost:
$0.000245
Provider:
SambaNova
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
“Bill, are you listening? The target is a seven-year-old’s birthday party in a bouncy castle fortress. Our intelligence suggests a perimeter guarded by overly-caffeinated parents.” “A bouncy castle? Molly, we’re thieves, not party clowns. Did you even case the joint?” “Of course I did. The main loot is under the cake table. The primary obstacle is a five-foot-tall ‘princess’ with a glitter wand who asks ‘why’ a lot. Distraction is key. I’ll create a diversion by ‘accidentally’ stepping on the birthday cake.” “You’ll what? No! We need something subtle. I’ll deploy Operation Sugar Rush. I’ll spike the lemonade with industrial-grade candy. The ensuing hyperactive chaos will be our cover.” “That’s… brilliant. But what about the dog? The family has a geriatric beagle named Mr. Wiggles.” “A beagle? Molly, that’s not a guard dog, it’s a furry tripod. I’ll bribe it with a squeaky toy. Now, the exit strategy: we escape through the side gate, loaded with presents, disguised as disgruntled party entertainers.” “Disgruntled clowns. Perfect. Our greatest heist yet. We’ll be legends.” “We’ll be arrested for grand theft lollipop. Just remember, no pinching the goody bags until we’re clear.”
created_at
"2026-02-16T21:29:12.366506+00:00"
2/16/2026, 9:29:12 PM
model
"deepseek/deepseek-chat-v3.1"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2066
moderation_latency
(null)
generation_time
21873
tokens_prompt
95
tokens_completion
338
native_tokens_prompt
67
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
0.0002448
router
(null)
provider_responses
0
id
"deb122a1-0b45-4c51-96d6-82c3564de59c"
status
200
is_byok
false
latency
2065
endpoint_id
"1dda4bb9-c626-4816-bb54-99665323fef0"
provider_name
"SambaNova"
model_permaslug
"deepseek/deepseek-chat-v3.1"
api_type
"completions"
id
"gen-1771277330-AZYr8xGwRmEeKxLAXBIu"
upstream_id
"deb122a1-0b45-4c51-96d6-82c3564de59c"
total_cost
0.0002448
cache_discount
(null)
upstream_inference_cost
0
provider_name
"SambaNova"
Evaluation details
Result
Evaluator
Details
Meta Data
99.3769%
Matches word count
n/a
neededClean
false
words
195
36.7879%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
195
wordsDialogue
195
68.0824%