NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5HY2RWCRMCNS6RGCJ03B4
Run Details
Model:
DeepSeek V3 (2025-03-24)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
11.6s
Cost:
$0.000342
Provider:
SiliconFlow
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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:** "Okay, hear me out. The Johnson kid’s birthday party. Saturday. Loads of presents, minimal security." **Bill:** "Minimal? You mean a bunch of sugar-crazed six-year-olds and at least one hyper-vigilant mom." **Molly:** "Exactly. Distractions *everywhere*. We dress as clowns, blend in, grab the loot while they’re busy with the piñata." **Bill:** "Clowns? I *hate* clowns." **Molly:** "That’s why it’s perfect. No one questions a clown at a kid’s party, and they won’t look *too* close because clowns are creepy." **Bill:** "You ever actually *been* a clown?" **Molly:** "How hard can it be? Big shoes, squeaky nose, steal stuff. We’re professionals." **Bill:** "What if someone asks for balloon animals?" **Molly:** "We hand them a deflated balloon and run. Priorities, Bill!" **Bill:** "Fine. But if a kid cries on me, I’m bolting." **Molly:** "Deal. Just remember—focus on the loot. No getting sucked into a game of musical chairs." **Bill:** "Oh, now you tell me." **Molly:** "And no eating the cake until *after* we’re clear." **Bill:** "*Somehow* that’s the hardest rule."
created_at
"2026-02-16T21:26:39.265785+00:00"
2/16/2026, 9:26:39 PM
model
"deepseek/deepseek-chat-v3-0324"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2932
moderation_latency
(null)
generation_time
11518
tokens_prompt
77
tokens_completion
311
native_tokens_prompt
66
native_tokens_completion
326
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
0.0003425
router
(null)
provider_responses
0
id
"019c6858f90fd6f0429e6da7e8384f48"
status
200
is_byok
false
latency
2932
endpoint_id
"d562908f-c5eb-4a32-9a91-24d53b176fd8"
provider_name
"SiliconFlow"
model_permaslug
"deepseek/deepseek-chat-v3-0324"
api_type
"completions"
id
"gen-1771277187-SghCrKQw2yuqpnWV3vfj"
upstream_id
"019c6858f90fd6f0429e6da7e8384f48"
total_cost
0.0003425
cache_discount
(null)
upstream_inference_cost
0
provider_name
"SiliconFlow"
Evaluation details
Result
Evaluator
Details
Meta Data
0.4920%
Matches word count
n/a
neededClean
false
words
173
99.8677%
Dialogue to Total Word Ratio
Ratio: 91.91%, Deviation: 1.91%
neededClean
false
wordsTotal
173
wordsDialogue
159
50.1799%