NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5GAMQSCTNJY2TP1QYMCFN
Run Details
Model:
DeepSeek V3 (2025-03-24)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
8.5s
Cost:
$0.000538
Provider:
Crusoe
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
Molly adjusted her fake clown nose and scowled. “Bill, this is the dumbest plan you’ve ever had, and that includes the time you tried to steal a dolphin.” Bill tightened the polka-dot suspenders of his oversized costume. “No, no, hear me out. Kids’ parties are *perfect*. The parents are distracted, the security’s non-existent, and the loot is all in one place—wrapped and ready. Plus, we look hilarious.” He did a wobbly cartwheel for emphasis. “We look like we escaped from a low-budget circus,” Molly muttered, eyeing her rainbow wig in a pocket mirror. “What if someone recognizes us?” “Recognize us? Molly, we’re wearing neon overalls and squeaky shoes. We’re invisible.” He grabbed a half-inflated balloon and twisted it into a lopsided sword. “I’ll keep the kids busy with balloon animals—bad ones—while you grab the presents. They’ll never suspect a thing.” Molly sighed. “I can’t believe I’m saying this… but fine. If we get caught, though, I’m blaming the clown makeup.” Bill grinned, smearing his face paint further. “That’s the spirit! Now, remember: if things go south, just scream ‘PINATA’ and make a run for it.” Molly pinched the bridge of her nose. “I hate my life.” “No, you hate *not* being rich,” Bill corrected. “Now let’s go rob some toddlers.”
created_at
"2026-02-16T21:25:43.534835+00:00"
2/16/2026, 9:25:43 PM
model
"deepseek/deepseek-chat-v3-0324"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
541
moderation_latency
(null)
generation_time
8443
tokens_prompt
77
tokens_completion
363
native_tokens_prompt
67
native_tokens_completion
336
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)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.0005375
router
(null)
provider_responses
0
id
"cmpl-a43aed1105669f82194338b3867f3081"
status
200
is_byok
false
latency
435
endpoint_id
"eb0625b2-17a1-492d-b3cc-e35091f1431d"
provider_name
"Crusoe"
model_permaslug
"deepseek/deepseek-chat-v3-0324"
api_type
"completions"
id
"gen-1771277134-a6BHtVxf64lELJingW9G"
upstream_id
"cmpl-a43aed1105669f82194338b3867f3081"
total_cost
0.0005375
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Crusoe"
Evaluation details
Result
Evaluator
Details
Meta Data
43.3783%
Matches word count
n/a
neededClean
false
words
217
0.0000%
Dialogue to Total Word Ratio
Ratio: 69.59%, Deviation: 19.59%
neededClean
false
wordsTotal
217
wordsDialogue
151
21.6892%