NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2EDNEY01A9KZH5Z153FMH5
Run Details
Model:
DeepSeek V4 Pro (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
11.2s
Cost:
$0.000435
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
86.38%
Matches word count
(details)
79.48%
Dialogue to Total Word Ratio
(details)
82.9280%
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 unfolded a crinkled birthday napkin across the sticky diner table, revealing a crayon-drawn floor plan of a bouncy castle’s inflation port. Bill leaned in, squinting at the wobbly lines that indicated a piñata blind spot. He tapped a calloused finger on the glitter-glued X marking the gift table, then mimed the decisive snip of bolt cutters through a twist-tie garland. Their whispered strategy involved a smoke machine and a highly trained ferret named Nicodemus. Bill produced a pair of animal-shaped novelty slippers, the rubber soles perfect for silent pizza-print linoleum. Molly practiced a harried parent frown, adjusting her oversized sunglasses, before sliding a party-blower through her buttonhole like a master safecracker’s tool. Bill grunted, “The clown is the real variable. We put the sedative in his seltzer bottle, or we’re juggling witnesses.” Outside, a minivan with a faded “Honk if You Love Unicorns” bumper sticker idled, its trunk already lined with empty toy chests. Molly nodded, her voice a gravelly murmur. “Chuck E. Cheese protocol. No fingerprints on the cupcake tier.” Their hearts beat in sync with the distant, percussive thump of an inflatable moonwalk reaching critical pressure.
created_at
"2026-04-25T13:49:15.618Z"
4/25/2026, 1:49:15 PM
model
"deepseek/deepseek-v4-pro-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
738
moderation_latency
(null)
generation_time
11106
tokens_prompt
77
tokens_completion
503
native_tokens_prompt
67
native_tokens_completion
466
native_tokens_completion_images
(null)
native_tokens_reasoning
182
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
0.000434565
router
(null)
provider_responses
0
endpoint_id
"d084170b-dd99-4e53-9ae1-b63881f3cfbf"
id
"7b5f02b7-918f-4ee7-97f5-d176255355b1"
is_byok
false
latency
286
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-1777124955-H7WdC2yEIw2OLtbIz6dP"
session_id
(null)
api_type
"completions"
id
"gen-1777124955-y8pQgftIGhnmwfx7UirU"
upstream_id
"7b5f02b7-918f-4ee7-97f5-d176255355b1"
total_cost
0.000434565
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
86.38%
Matches word count
n/a
neededClean
false
words
189
79.48%
Dialogue to Total Word Ratio
Ratio: 16.92%, Deviation: 6.92%
neededClean
false
wordsTotal
195
wordsDialogue
33
82.9280%