NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1K9ME2RE01VZHGEE25HS6H1
Run Details
Model:
DeepSeek V4 Pro 0813 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
17.2s
Cost:
$0.007290
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 70% 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 squinted at the crayon-drawn map. "Okay, run it again. We hit the bouncy castle at exactly 2:15, after the magician starts but before the piñata. Why before the piñata?" Bill adjusted his party hat. "Because after the piñata, the sugar kicks in. You ever tried to outrun a six-year-old on a juice-box high? It's like chasing a greased weasel." "Fair. So we enter as what, exactly? Clowns?" "Worse. Entertainment coordinators. I've got lanyards." "Lanyards? Bill, it's a child's birthday party, not a corporate retreat." "Trust me. Parents see a lanyard, they assume you're in charge. We walk in, say we're from Pony Time Productions, and we're here to inspect the gift table for safety compliance." Molly snorted. "Safety compliance? For a pile of wrapped boxes?" "Choking hazards, unstable wrapping paper, ribbon-related injuries. It's a real thing. I printed a clipboard with a fake checklist." "You printed a clipboard?" "I printed a checklist and glued it to a clipboard. It says Gift Table Risk Assessment. I even laminated it." Molly stared. "You laminated it." "Professionalism, Molly. It disarms people. Now, while I distract the parents with a speech about gift-bag suffocation, you slip behind the inflatable unicorn and start loading presents into the diaper bag." "Why a diaper bag?" "Because no one questions a man carrying a diaper bag. It's the perfect disguise. It says I have given up on life, but I still love my child." "You don't have a child." "The bag doesn't know that." Molly tapped the map. "What about the birthday girl? She's going to notice her mountain of presents shrinking." Bill waved a hand. "I've got a decoy. A box of empty cardboard wrapped in shiny paper. We replace the real presents with decoys as we go. By the time she opens them, we're three towns away." "And when she opens a box full of air?" "She'll think it's a metaphor. Kids love abstract concepts. Or she'll cry. Either way, we're gone." Molly frowned. "What about the clown? You said there's a clown." "Ah, yes. Bubbles the Clown. He's a wildcard. He makes balloon animals and smells like regret. We avoid eye contact. If he offers you a balloon poodle, you take it and keep moving." "Why would I take it?" "Because refusing a balloon poodle is suspicious. Everyone loves a balloon poodle." Molly sighed. "Fine. I take the poodle. Then what?" "Then you say, Oh, I left my safety whistle in the van, and you walk out with the diaper bag full of presents. I'll create a diversion by pretending to choke on a cupcake." "You're going to fake choke?" "It's the only way to clear a room of parents. They'll all rush to perform the Heimlich, and no one will watch the gift table. It's flawless." "Except you might actually choke." "I've practiced. I can cough on command. Watch." Bill coughed weakly. "See? Tragic, yet non-lethal." Molly folded the map. "This is the dumbest plan I've ever heard." Bill grinned. "Which is why it'll work. Nobody suspects idiots." "That's not the compliment you think it is." "Are you in or not?" Molly grabbed the diaper bag. "I'm in. But if that clown tries to make me a giraffe, I'm out."
created_at
"2026-09-03T09:29:38.149Z"
9/3/2026, 9:29:38 AM
model
"deepseek/deepseek-v4-pro-20260813"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
594
moderation_latency
(null)
generation_time
16930
tokens_prompt
75
tokens_completion
1867
native_tokens_prompt
147
native_tokens_completion
1792
native_tokens_completion_images
(null)
native_tokens_reasoning
1011
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
0.00729036
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788427778-I2kToV3dv0qOjJ7HvdT7"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788427778-wjf1CcS72nJ4Qtr6BTUp"
upstream_id
"chatcmpl-25d96409fdf74be18cebe5d1f5cdc72b"
provider_responses
0
endpoint_id
"85bb5a4a-983d-4d11-b97b-c7e5b348375f"
id
"chatcmpl-25d96409fdf74be18cebe5d1f5cdc72b"
is_byok
false
latency
594
model_permaslug
"deepseek/deepseek-v4-pro-20260813"
provider_name
"BaseTen"
status
200
total_cost
0.00729036
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
537
0.00%
Dialogue to Total Word Ratio
Ratio: 92.28%, Deviation: 22.28%
neededClean
false
wordsTotal
544
wordsDialogue
502
0.0000%