NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KB5R886YP84MQWMFEYWC40
Run Details
Model:
DeepSeek V4 Pro 0813 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
33.8s
Cost:
$0.011381
Provider:
Sail Research
Evaluation summary
Score
Evaluator
Details Link
99.92%
Matches word count
(details)
99.67%
Dialogue to Total Word Ratio
(details)
99.7928%
Overall
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
In a van parked across from the bounce house, Bill unfolded a crudely drawn map and tapped the gift table. “You distract the clown, I’ll grab the gift table,” Molly whispered, adjusting a pair of oversized sunglasses. “Why do I always get the clown?” Bill asked, wiping sweat from his brow. “Because you still have that rainbow wig from the bank job.” He groaned and stared at the inflatable slide. “Those kids are vicious. Last time I tried to take a juice box, a four-year-old bit my ankle.” Molly patted his arm with a latex-gloved hand. “That’s why we go in during the piñata. They’ll be blindfolded and swinging.” Outside, a minivan full of balloons pulled up and a dozen small guests spilled out. “Remember, no real names,” Molly said, handing him a lollipop. “You’re Captain Wobbles. I’m Aunt Sparkles.” Bill frowned at his sequined vest. “Can I be Uncle Glitter?” “No, Uncle Glitter is wanted in two states.” She adjusted her fake mustache and checked the GPS tracker hidden in a party hat. “Ready?” Bill stuffed a pillowcase into his waistband. “If I die, delete my browser history.” “Deal.” They fist-bumped and cracked the van door. “Let’s go steal some crayon-covered gold.”
created_at
"2026-09-03T09:56:34.201Z"
9/3/2026, 9:56:34 AM
model
"deepseek/deepseek-v4-pro-20260813"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
799
moderation_latency
(null)
generation_time
33674
tokens_prompt
75
tokens_completion
2474
native_tokens_prompt
146
native_tokens_completion
2872
native_tokens_completion_images
(null)
native_tokens_reasoning
2577
native_tokens_cached
145
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"
service_tier
(null)
usage
0.01138082
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788429394-4E4VpAjYKQiNYW8IQ6WL"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788429394-HOp4mPdLkZ3D0Ko20biG"
upstream_id
"asapreq_aaaf6a30d1d914c9c6dfc43b0bbd73e5"
provider_responses
0
endpoint_id
"6811c06b-8624-4796-ad6e-807e33b97ae6"
id
"asapreq_aaaf6a30d1d914c9c6dfc43b0bbd73e5"
is_byok
false
latency
799
model_permaslug
"deepseek/deepseek-v4-pro-20260813"
provider_name
"Sail Research"
status
200
total_cost
0.01138082
cache_discount
0.00018502
upstream_inference_cost
0
provider_name
"Sail Research"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
99.92%
Matches word count
n/a
neededClean
false
words
203
99.67%
Dialogue to Total Word Ratio
Ratio: 47.60%, Deviation: 2.40%
neededClean
false
wordsTotal
208
wordsDialogue
99
99.7928%