NC Bench
Overview
About
Tests
Models
Model
Mistral Medium
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-07-19-01J35WQA79KAG3503FKFHV4G6T
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: The kiddie bash is at 3 PM sharp, we've got to move fast. You got the costumes? Bill: Yep, clown suits. Ironic, ain't it? Stealing presents while dressed as a children's entertainer. Molly: Just stick to the plan. We blend in, grab the presents, and bolt. Remember, the birthday boy's parents are filthy rich. These gifts are probably worth a fortune. Bill: Easy peasy. But how do we know which presents are the expensive ones? Molly: That's why we're bringing the scanner. It'll detect high-value items. Just remember to keep your mask on. Kids have sharp eyes and freaky memories. Bill: Won't the parents think it's weird if two clowns are scanning gifts? Molly: We'll tell them it's a new party game. C'mon Bill, use your charm. You can sweet-talk your way out of anything. Bill: You know it. Say, have you ever noticed how kids' parties have the best food? Molly: Focus, Bill! We're not here for the vol-au-vents, we're here for the loot. Bill: Right, right. Just can't resist a good puff pastry.
id
"gen-UYbfpziX9RSQX8GArswQYSsZGvF4"
total_cost
0.0028701
upstream_id
"eecab0f50c384268a4581612c0021e6f"
created_at
"2024-07-19T16:20:53.640517+00:00"
Friday, July 19th, 2024 at 4:20:53 PM GMT+00:00
model
"mistralai/mistral-medium"
app_id
182717
streamed
false
cancelled
false
provider_name
"Mistral"
latency
8141
moderation_latency
(null)
generation_time
0
finish_reason
"stop"
tokens_prompt
68
tokens_completion
258
native_tokens_prompt
85
native_tokens_completion
326
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.0028701
Evaluation details
Result
Evaluator
Details
Meta Data
20.1897%
Matches word count
n/a
neededClean
false
words
180
0.0000%
Dialogue to Total Word Ratio
Ratio: 3.33%, Deviation: 86.67%
neededClean
false
wordsTotal
180
wordsDialogue
6
10.0948%