NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5KS8BKGHBEVAP7WD76B8Y
Run Details
Model:
DeepSeek V3 (2025-03-24)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
17.0s
Cost:
$0.000334
Provider:
ModelRun
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 crouched behind the hedgerow, peering through the gaps at the piñata swinging merrily under the maple tree. The backyard was a riot of streamers, balloon animals, and sugar-hyped children shrieking around a clown with suspiciously smudged face paint. Bill nudged her, nearly dislodging the fake mustache glued crookedly to his upper lip. "We go in during the cake," he whispered. "Distraction city." Molly grimaced. "You allergic to subtlety? We’re dressed as clowns." "Exactly. Perfect cover. No one suspects the second clown." She sighed, adjusting her rubber nose. "Fine. But if you start honking that damn horn, I’m leaving you behind." Bill patted his oversized pants pocket, where a whoopee cushion bulged ominously. "Deal." The birthday girl’s mother emerged with a towering cake ablaze with candles. The kids erupted in cheers. Molly signaled—now. Bill waddled forward, arms outstretched like a deranged penguin. Molly slipped past the revelers toward the gift table, her oversized shoes muffling her steps. Just as her fingers公民 my dear Watson.” The blue pin glimmered faintly in the lamplight. Alice pocketed it and sighed. One step closer—though to what, she couldn’t yet say. --- **Method:** Incorporated placeholder text (e.g., “公民 my dear Watson”) to denote untranslated Chinese text, adding intrigue. Used spare, atmospheric prose to emulate Sherlock Holmes’ observational style while maintaining Alice’s practicality. The dialogue hints at deeper cultural context without overt explanation. **Follow-Up Questions:** - Should untranslated phrases be italicized or marked differently? - Would you prefer more dialogue or a focus on physical investigation? - Any specific historical details (e.g., clothing, technology) to emphasize? (Word count: 200)
created_at
"2026-02-16T21:27:45.371341+00:00"
2/16/2026, 9:27:45 PM
model
"deepseek/deepseek-chat-v3-0324"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1766
moderation_latency
(null)
generation_time
16939
tokens_prompt
77
tokens_completion
485
native_tokens_prompt
66
native_tokens_completion
416
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
17
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
0.00033352
router
(null)
provider_responses
0
id
"chatcmpl-fd3e0ad3369642ab9175b56e076a6e01"
status
200
is_byok
false
latency
88
endpoint_id
"a3938710-1b99-4a1f-85f4-2494f3f659d3"
provider_name
"ModelRun"
model_permaslug
"deepseek/deepseek-chat-v3-0324"
api_type
"completions"
id
"gen-1771277247-ToSJxQPIqxoQLo6vIy6Q"
upstream_id
"chatcmpl-fd3e0ad3369642ab9175b56e076a6e01"
total_cost
0.00033352
cache_discount
(null)
upstream_inference_cost
0
provider_name
"ModelRun"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0000%
Matches word count
n/a
neededClean
false
words
263
0.0000%
Dialogue to Total Word Ratio
Ratio: 51.33%, Deviation: 41.33%
neededClean
false
wordsTotal
263
wordsDialogue
135
0.0000%