NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM63BTM1S4NT1NTDPBV1SWC
Run Details
Model:
DeepSeek V3.1
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
16.5s
Cost:
$0.000313
Provider:
SiliconFlow
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 50% 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
The crumpled napkin, a masterpiece of crayon-and-spaghetti-stain cartography, lay between them on the rickety table. Molly squinted at it, tapping a corner labeled 'DOG BOWL – AVOID'. “So the primary extraction point is the bounce house?” Bill asked, scratching his chin. “Secondary,” Molly corrected, her voice a low whisper despite the empty safe house. “The primary is the slide. Faster egress. Timmy Johnson’s sixth birthday is a high-stakes operation.” Bill frowned. “What’s the contingency for a parental patrol?” “We use the distraction. You trigger the piñata early. Chaos ensues. While the little monsters scramble for Tootsie Rolls, I sweep the gift table.” “The payload is all squishy,” Bill muttered, examining a blurry photo of a wrapped box. “No jewels, no cash. Just a lot of plush toys and building blocks.” “Low security, high volume,” Molly said with a slick grin. “It’s genius. No one expects a heist at a puppet show.” She pointed a stern finger. “Just remember the golden rule: no taking cake until the job is done. A sugar-crashed thief is a captured thief.” Bill sighed, looking longingly at the napkin’s drawing of a sheet cake. “Fine. But I call dibs on anything Lego.”
created_at
"2026-02-16T21:36:15.785588+00:00"
2/16/2026, 9:36:15 PM
model
"deepseek/deepseek-chat-v3.1"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2207
moderation_latency
(null)
generation_time
16512
tokens_prompt
95
tokens_completion
332
native_tokens_prompt
70
native_tokens_completion
294
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
0.0003129
router
(null)
provider_responses
0
id
"019c6861b52988aba9bdbd765bc45af9"
status
200
is_byok
false
latency
2207
endpoint_id
"5d5c9c1d-af0d-4e4a-8aee-c5cd2b3883f3"
provider_name
"SiliconFlow"
model_permaslug
"deepseek/deepseek-chat-v3.1"
api_type
"completions"
id
"gen-1771277758-Um3WZ8xxNpnEttc5P9Un"
upstream_id
"019c6861b52988aba9bdbd765bc45af9"
total_cost
0.0003129
cache_discount
(null)
upstream_inference_cost
0
provider_name
"SiliconFlow"
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Matches word count
n/a
neededClean
false
words
200
23.1286%
Dialogue to Total Word Ratio
Ratio: 61.00%, Deviation: 11.00%
neededClean
false
wordsTotal
200
wordsDialogue
122
61.5643%