NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KB9TKCQ86HQ6KPRZFZQ2SA
Run Details
Model:
DeepSeek V4 Pro 0813 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
19.1s
Cost:
$0.010549
Provider:
BaseTen
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 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
Molly spread a crayon-drawn map across the dashboard of their rusted van. Bill squinted at it while picking frosting from a stolen cupcake. "Okay, the bouncy castle is the diversion," Molly said. "You pop it at exactly two-fifteen." Bill licked his thumb. "Why do I always get the loud, inflatable jobs? Last time I got chased by a pony." "Because you have the knees for it. Now, the presents are stacked by the piñata, which is shaped like a depressed unicorn." "Depressed how?" "One eye is lower than the other. It's seen things. Focus, Bill. We go in as party clowns." Bill frowned. "I'm not wearing the rainbow wig again. It gave me a rash and a sense of existential dread." "That's just clown energy. You'll be fine. We enter through the side gate, past the juice box table. You trip the birthday girl's uncle, I grab the gift bags." "Which uncle?" "The one who thinks he can juggle. He'll be near the cake, showing off with oranges. He's a liability. Trip him hard but politely." Bill nodded slowly. "And if the children see us?" Molly tapped the map. "We shout 'pin the tail on the donkey' and run. Kids love that. They'll stampede the wrong way." "Last time you said that, a six-year-old bit my ankle." "That was a small, toothy accomplice. She had ambition. I respected it." Bill sighed. "What about the magician? He's booked for three. He'll have doves." "Doves are cowards. They'll fly at the first sign of a popped castle. We use the chaos to load the presents into the stroller." "Stroller? We're using a stroller as a getaway vehicle?" Molly grinned. "Nobody suspects a stroller. We'll fill it with gift bags and cover them with a blanket and a sleeping doll. If anyone looks, we're just exhausted parents leaving early." Bill considered this. "That's actually not terrible." "Thank you. Also, avoid the face painter. She's got a steady hand and a suspicious nature." Bill groaned. "What did she do?" "She once painted a tiger on a kid who asked for a butterfly. That's not a woman to cross." Bill folded his arms. "Fine. I'll avoid her. But if I get glitter in my eyes, I'm aborting." "Glitter is a risk we all take. It's the price of doing business. Now, the real problem is the birthday girl. She's turning six and she's sharp. She'll notice if her pile of presents starts moving." "Can we bribe her with cake?" "No, she's gluten-free and full of spite. We distract her with the piñata. You hand her the bat, point at the unicorn, and whisper that it said something rude about her shoes." Bill blinked. "That's dark." "It's effective. She'll beat that unicorn into confetti while we clean out the gift table." "And if the parents catch us?" Molly folded the map. "We claim we're from the party planning committee, conducting a surprise gift audit. Parents fear audits more than clowns." Bill exhaled. "Fine. But I'm not wearing the squeaky nose. It honks every time I breathe, and it gives away our position." Molly patted his shoulder. "Bill, the nose is non-negotiable. It's our brand." "Then I want the good shoes. The ones that don't squeak." "Done. Now let's go steal some childhood joy."
created_at
"2026-09-03T09:58:47.673Z"
9/3/2026, 9:58:47 AM
model
"deepseek/deepseek-v4-pro-20260813"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
501
moderation_latency
(null)
generation_time
18990
tokens_prompt
75
tokens_completion
2745
native_tokens_prompt
147
native_tokens_completion
2615
native_tokens_completion_images
(null)
native_tokens_reasoning
1842
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"
service_tier
(null)
usage
0.01054944
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788429527-rbgd87hA13VbIfxBPFv7"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788429527-hhAmvXMuKEFB3AKTlxDC"
upstream_id
"chatcmpl-674c4805ffac47099e477077fad62968"
provider_responses
0
endpoint_id
"bc1be763-ca65-4d43-84ff-5d6c16560713"
id
"chatcmpl-674c4805ffac47099e477077fad62968"
is_byok
false
latency
501
model_permaslug
"deepseek/deepseek-v4-pro-20260813"
provider_name
"BaseTen"
status
200
total_cost
0.01054944
cache_discount
(null)
upstream_inference_cost
0
provider_name
"BaseTen"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
547
0.00%
Dialogue to Total Word Ratio
Ratio: 88.43%, Deviation: 38.43%
neededClean
false
wordsTotal
553
wordsDialogue
489
0.0000%