NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM626JF0XG3R6XVSDZGEPEA
Run Details
Model:
DeepSeek V3.1
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
8.5s
Cost:
$0.000338
Provider:
Novita
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
"So it's a solid plan," Molly said, peering over the blueprint of the suburban house. "We go in disguised as the hired Princess Sparklehoof and Sir Gigglebottom." Bill adjusted his fake beard. "The what now?" "The clown and the unicorn, Bill. Keep up. We infiltrate during the peak chaos of the piƱata assault. I'll cause a distraction with a glitter bomb." "A glitter bomb? That's messy." "It's tactical, Bill! While the tiny humans are blinded by shimmer, you sweep the gift table. In and out. They'll be none the wiser." Bill frowned, tapping the diagram. "What about the perimeter? We've got a dozen small, hyper-vigilant witnesses and at least one overly-friendly golden retriever." "The dog is a non-issue. I've laced the birthday cake with a mild sedative. He'll be snoring by the presents." "You poisoned the cake?" "It's a veterinarian-grade relaxant! Barely a nibble will have Fido dreaming of squirrels. The real problem is extraction. The Getaway Dad-Van only fits so many brightly wrapped packages." "We'll prioritize. Big boxes, small boxes. No gift cards." "Agreed. Too much paperwork. Are we ready? Do you have your whoopee cushion?" Bill patted his jester costume. "Security blanket. Remember, if things go south, the code phrase is 'I think little Timmy needs a diaper change.'" "Perfect. Let's go make some children cry."
created_at
"2026-02-16T21:35:29.064147+00:00"
2/16/2026, 9:35:29 PM
model
"deepseek/deepseek-chat-v3.1"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
802
moderation_latency
(null)
generation_time
8384
tokens_prompt
95
tokens_completion
357
native_tokens_prompt
68
native_tokens_completion
320
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
0.00033836
router
(null)
provider_responses
0
id
"6d89f035-7965-4e9c-b38d-d509f415aa89"
status
200
is_byok
false
latency
802
endpoint_id
"715d05e1-9fa7-4ceb-834a-833f5e26a55a"
provider_name
"Novita"
model_permaslug
"deepseek/deepseek-chat-v3.1"
api_type
"completions"
id
"gen-1771277720-z8znrH5Hri2l6mmr4d6L"
upstream_id
"6d89f035-7965-4e9c-b38d-d509f415aa89"
total_cost
0.00033836
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Novita"
Evaluation details
Result
Evaluator
Details
Meta Data
6.0907%
Matches word count
n/a
neededClean
false
words
223
99.9785%
Dialogue to Total Word Ratio
Ratio: 88.79%, Deviation: 1.21%
neededClean
false
wordsTotal
223
wordsDialogue
198
53.0346%