NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2HCS70HFB4DE4T0AB2KNYG
Run Details
Model:
DeepSeek V4 Pro (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
9m 3s
Cost:
$0.041873
Provider:
Together
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
42.05%
Dialogue to Total Word Ratio
(details)
21.0232%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 and Bill were seasoned thieves, but this was a first: a children’s party. The van, a rusted blue relic from the eighties, smelled of coffee and regret as it shuddered to a halt across from the house. The front lawn was a riot of color—streamers twisted around trees, a bouncy castle inflating in the back, and a wobbly sign that read “Happy Birthday, Timmy!” Through the window, Molly spotted the mountain of gift-wrapped boxes, their shiny surfaces a siren call. Balloons shaped like cartoon characters bobbed in the windows, and the sound of children’s laughter spilled through the open door. The smell of sugar and burnt popcorn wafted their way. This was the perfect mark. “We need a plan,” Bill said, his eyes darting from the house to Molly. Molly nodded, adjusting her earpiece with a practiced flick. “Something simple. In and out. Thirty seconds.” Bill leaned forward, waving his hands in the air. “But where’s the style? We’re artists, not common thieves.” She raised an eyebrow, her lips twitching. “What’s your brilliant idea? A musical number?” “Better. We disguise ourselves as the presents themselves. Giant boxes, all wrapped up. We mingle with the other gifts.” He grinned, proud of his ingenuity, and tapped a finger on his temple. Molly laughed, a sharp, barking sound. “And how do we move? Hop across the room?” “Wheels on the bottom. Remote controlled. I seen it in a cartoon once.” Bill crossed his arms, his leather jacket creaking, daring her to challenge the genius. “That’s your source? A cartoon? What’s next, a tunnel under the sandbox?” She sighed and rubbed her temples, her nails chipped from the last job. The van’s dashboard clock ticked loudly, a relentless reminder. Bill’s shoulders slumped. He toyed with a stray thread on his jacket, the bravado draining. “Fine, but I’m keeping the idea for later. Maybe a bank job.” “Sure, Bill.” Molly checked her watch and slid a ski mask over her face, the wool itching her cheeks. “Now, let’s move. And try not to trip over the toddlers.” They exchanged a look, a mix of determination and the usual pre-heist jitters. The van door creaked open, and they slinked toward the house, the laughter growing louder with each step. Molly’s boots crunched on the gravel path, while Bill adjusted his fake mustache, which immediately drooped on one side. Inside, a child shrieked with delight, and a piñata cracked open, showering candy. This birthday was about to get a lot less festive.
created_at
"2026-04-25T14:41:14.631Z"
4/25/2026, 2:41:14 PM
model
"deepseek/deepseek-v4-pro-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
41382
moderation_latency
(null)
generation_time
540689
tokens_prompt
77
tokens_completion
8962
native_tokens_prompt
146
native_tokens_completion
9447
native_tokens_completion_images
(null)
native_tokens_reasoning
8299
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.0418734
router
(null)
provider_responses
0
endpoint_id
"d084170b-dd99-4e53-9ae1-b63881f3cfbf"
is_byok
false
latency
280
model_permaslug
"deepseek/deepseek-v4-pro-20260423"
provider_name
"DeepSeek"
status
429
1
endpoint_id
"13bda129-f8a6-4a60-b1db-98082e5f7ccd"
id
"ogZiuDM-6z2FuN-9f1e1b846591e865"
is_byok
false
latency
41382
model_permaslug
"deepseek/deepseek-v4-pro-20260423"
provider_name
"Together"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777128074-lU8glgxnGqi348VGt95A"
session_id
(null)
api_type
"completions"
id
"gen-1777128074-LoZ8JK0tc5PooKORLVt2"
upstream_id
"ogZiuDM-6z2FuN-9f1e1b846591e865"
total_cost
0.0418734
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Together"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
417
42.05%
Dialogue to Total Word Ratio
Ratio: 25.84%, Deviation: 4.16%
neededClean
false
wordsTotal
418
wordsDialogue
108
21.0232%