NC Bench
Overview
About
Tests
Models
Model
DeepSeek-V2 Chat
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-07-19-01J35SWTD941DF8NMK6172RRRQ
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: "So, Bill, you ever think about how weird it is that we're about to rob a bunch of kids?" Bill: "Only every time I look at the invite. 'Princess Penelope's Magical Birthday Bash.' We're the worst." Molly: "Hey, it's not like we're taking their candy. Just the presents. Big difference." Bill: "Right, because candy's a gateway drug. Presents are the gateway to a life of luxury." Molly: "Exactly! And think about it, no security. Just a bunch of parents too busy Instagramming their kid's cake-smashing to notice us." Bill: "True. And the presents are probably all unmarked, so we can pawn them off without a hitch." Molly: "See? We're practically philanthropists. We're redistributing wealth from the hands of doting relatives to the pawn shop economy." Bill: "Redistributing wrapped boxes, one at a time. I feel like Robin Hood, but with more bows and less arrows." Molly: "And less moral ambiguity. We're doing this for the greater good. Of our bank accounts." Bill: "Here's to the greater good. And to not getting caught by a bunch of toddlers." Molly: "Cheers to that. Now, let's go steal some joy."
id
"gen-UbaG4cEw5caT6CfUgetZllOwc5Ut"
total_cost
0.00009646
upstream_id
"189324c07309e50d3b3ff9237a9a44b5"
created_at
"2024-07-19T15:31:32.917411+00:00"
Friday, July 19th, 2024 at 3:31:32 PM GMT+00:00
model
"deepseek/deepseek-chat"
app_id
182717
streamed
false
cancelled
false
provider_name
"DeepSeek"
latency
17639
moderation_latency
(null)
generation_time
0
finish_reason
"stop"
tokens_prompt
68
tokens_completion
276
native_tokens_prompt
73
native_tokens_completion
308
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.00009646
Evaluation details
Result
Evaluator
Details
Meta Data
86.3803%
Matches word count
n/a
neededClean
false
words
189
96.9936%
Dialogue to Total Word Ratio
Ratio: 94.18%, Deviation: 4.18%
neededClean
false
wordsTotal
189
wordsDialogue
178
91.6870%