NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2HM4A833TSMGZ3JFJY0AY3
Run Details
Model:
DeepSeek V4 Flash
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
18.7s
Cost:
$0.000177
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
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
The warehouse smelled of stale coffee and bad decisions. Molly was perched on a stolen tricycle, her legs too long for it, her expression one of profound skepticism as she examined the list in her gloved hand. Bill was crouched by a pile of clear plastic wrapping, muttering to himself. “So, the target is a ‘Princess Aurora’s Sparkle-Pony Soiree’ at the community center. Security is a single fourteen-year-old named Kyle who’s more interested in his phone than the door.” “I’m stuck on the moral quandary, Bill,” Molly said, spinning the trike’s squeaky front wheel. “Stealing presents from kids. It feels low, even for us.” “It’s not stealing, it’s redistribution. The parents spent too much. We’re teaching a lesson in fiscal responsibility.” He stood up, holding a roll of rainbow duct tape. “Plus, the haul is primo. The mom posted the registry online. We’re looking at four ‘Laser Blaster 5000s,’ two ‘Fart-in-a-Jar’ science kits, and a ‘My First Billionaire’ monopoly set that’s worth more than my car.” Molly sighed. “Right. Because nothing says ‘childhood magic’ like a depressed adult stealing a toy that farts.” “We don’t steal the magic. We steal the plastic. There’s a difference.” Bill unrolled a diagram of the center’s layout. “We go in through the kitchen. The cake is a decoy. I’ll create a diversion by setting off the ‘Fart-in-a-Jar’ kit near the punch bowl.” “You’re going to weaponize a science kit that teaches children about methane?” “It’s a proven distraction. Chaos is our friend. While chaos reigns, we swap the gift table with this decoy one I built.” He gestured to a sheet-covered lump that looked suspiciously like a card table holding several wrapped bricks. “We’re in and out in thirty seconds. They’ll be too busy gagging to notice.” Molly kicked a pedal. “And if the birthday girl, Princess Aurora herself, looks up with those big, un-blinky eyes and asks ‘Pwitty wady, why you take my dwagon?’” Bill paused. He thought for a very long second. “Then you don’t make eye contact. And you leave a note.” “A note?” Molly said. “A very clear, very professional I.O.U. for a replacement dwagon,” he said, his voice flat. “It’s called being a professional.” “We are not professional dwagon borrowers.” “We are whatever the job requires,” Bill said, grabbing a duffel bag. “Now stop philosophizing and start practicing your ‘responsible adult’ face. We have a kiddie bash to crash.”
created_at
"2026-04-25T14:45:13.164Z"
4/25/2026, 2:45:13 PM
model
"deepseek/deepseek-v4-flash-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
543
moderation_latency
(null)
generation_time
8505
tokens_prompt
77
tokens_completion
681
native_tokens_prompt
67
native_tokens_completion
600
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)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.00017738
router
(null)
provider_responses
0
endpoint_id
"934a69f9-bd54-474b-beca-24560f721e12"
is_byok
false
latency
59
model_permaslug
"deepseek/deepseek-v4-flash-20260423"
provider_name
"DeepInfra"
status
(null)
1
endpoint_id
"722e0746-2034-4d6e-a08f-e6271849ee9e"
id
"e744fae4-8dcd-4a4f-9b2a-4e42f365bf31"
is_byok
false
latency
288
model_permaslug
"deepseek/deepseek-v4-flash-20260423"
provider_name
"DeepSeek"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777128313-Oeqk3nJhRdoFVF2diWqJ"
session_id
(null)
api_type
"completions"
id
"gen-1777128313-WBDZzWRh28QRtvvULKMD"
upstream_id
"e744fae4-8dcd-4a4f-9b2a-4e42f365bf31"
total_cost
0.00017738
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
400
0.00%
Dialogue to Total Word Ratio
Ratio: 69.80%, Deviation: 39.80%
neededClean
false
wordsTotal
404
wordsDialogue
282
0.0000%