NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5T28ZW7RZ00RV601T3GQF
Run Details
Model:
DeepSeek V3.1
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
38.7s
Cost:
$0.000245
Provider:
DeepInfra
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 pudgy finger across the crudely drawn map, her brow furrowed in concentration. “The ‘vault’ is here. Heavily guarded by a sugar-crazed six-year-old named Timmy.” Bill adjusted his black beanie, peering over her shoulder at the diagram of the suburban backyard. “The big guy? The one with the plastic sword?” “The very same. Our intel suggests he’s ruthless. Took down a piñata in three swings.” They were huddled in their ‘hideout’—a dusty garden shed smelling of fertilizer and regret. The plan was simple, audacious: infiltrate the birthday party of one Emily Peterson and liberate the mountain of brightly wrapped gifts currently sitting unattended near the inflatable slide. “Distraction?” Bill whispered, his voice tight. Molly held up a whoopee cushion. “I’ll create a diversion at the snack table. You go in, grab the loot, and exfiltrate via the bouncy castle. It’s foolproof.” Bill eyed the flimsy whoopee cushion with deep skepticism. “That’s the entire plan? A whoopee cushion?” “It’s a classic, Bill. Never fails.” He sighed, accepting his fate. “Fine. But if we get caught, I’m telling your mother we’re here for the clown.”
created_at
"2026-02-16T21:31:33.087311+00:00"
2/16/2026, 9:31:33 PM
model
"deepseek/deepseek-chat-v3.1"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2094
moderation_latency
(null)
generation_time
38573
tokens_prompt
95
tokens_completion
324
native_tokens_prompt
67
native_tokens_completion
293
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
5
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
0.000245140000001
router
(null)
provider_responses
0
id
"cmpl-68c041ae8cd641b6b49d98985e8cca0d"
status
200
is_byok
false
latency
49
endpoint_id
"012f7d1f-23d4-4d03-9aa3-432d8ad82757"
provider_name
"DeepInfra"
model_permaslug
"deepseek/deepseek-chat-v3.1"
api_type
"completions"
id
"gen-1771277453-ue7pBcGWTafHsW8JFfNB"
upstream_id
"cmpl-68c041ae8cd641b6b49d98985e8cca0d"
total_cost
0.000245140000001
cache_discount
3.99999999e-7
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
81.2727%
Matches word count
n/a
neededClean
false
words
188
0.0000%
Dialogue to Total Word Ratio
Ratio: 52.66%, Deviation: 42.66%
neededClean
false
wordsTotal
188
wordsDialogue
99
40.6364%