NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2ETFJVPDKQ0QAE9Z8DB879
Run Details
Model:
DeepSeek V4 Pro
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
20.8s
Cost:
$0.000323
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
3.62%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
1.8117%
Overall
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 adjusted her binoculars, the lenses fogging slightly against the oppressive humidity of the rented chipmunk costume. Through the park’s wrought-iron fence, the target’s inflatable bouncy castle loomed, a garish monument to vulnerable sugar-high children. Beside her, Bill was methodically covering his reconnaissance sketch of the piñata with what looked suspiciously like a glitter-glue tactical strike map. A faint, tinny rendition of a song about a baby shark drifted from the clown-manned PA system. “I’m just saying, nobody body-searches a dancing squirrel,” Molly muttered, her voice muffled by faux fur. “Noted,” Bill replied, licking a dab of frosting from his lock-pick kit. “But are you sure the gift table isn’t a trap? That’s a professionally tiered cupcake display. It screams ‘personal security detail.’” “Those are the cake-a-razzi, Bill. Stay focused.” She pointed a furry paw towards a distracted father whose face was being aggressively painted like a tiger. “We infiltrate during the magic show. Classic misdirection. You create a diversion with a rogue juice box geyser, I’ll load the haul into a modified stroller.” Bill chewed his lip. “And the silent alarm?” “The birthday boy’s grandmother. But she’s a heavy sleeper after two glasses of Pinot Grigio.” Molly produced a single, expertly-wrapped decoy present, a box emitting a faint, anxious mewling sound. “She’ll be pinned down in the petting zoo for ten minutes, minimum.”
created_at
"2026-04-25T13:56:26.76Z"
4/25/2026, 1:56:26 PM
model
"deepseek/deepseek-v4-pro-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1025
moderation_latency
(null)
generation_time
9588
tokens_prompt
77
tokens_completion
394
native_tokens_prompt
67
native_tokens_completion
338
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
0.000323205
router
(null)
provider_responses
0
endpoint_id
"d084170b-dd99-4e53-9ae1-b63881f3cfbf"
id
"f2765936-e8ec-4468-bcc0-6adc24948ef2"
is_byok
false
latency
290
model_permaslug
"deepseek/deepseek-v4-pro-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-1777125386-H072yLFS2v2gTFEfvRL0"
session_id
(null)
api_type
"completions"
id
"gen-1777125386-Bpz4FPkk4DMbAm4aqZEr"
upstream_id
"f2765936-e8ec-4468-bcc0-6adc24948ef2"
total_cost
0.000323205
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
3.62%
Matches word count
n/a
neededClean
false
words
224
0.00%
Dialogue to Total Word Ratio
Ratio: 41.63%, Deviation: 31.63%
neededClean
false
wordsTotal
233
wordsDialogue
97
1.8117%