NC Bench
Overview
About
Tests
Models
Model
Magnum 72B
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-07-19-01J35T2W6BW064S1W3V2WHAEWK
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
"I still don't get it," Bill said. "We're professional burglars, not party crashers. Why can't we just rob a bank like normal crooks?" "Because banks have alarms, cameras, and guards. This way, we catch 'em with their pants down - no offense," Molly replied, double-checking her walkie-talkie. "We just need to time it right. Wait 'til Saint Nick is stuffing the chimney." "Right, and while he's doing that, I'll be hoisting myself up on a roof in a Santa suit! No thanks." Bill sighed heavily. "Cheer up. All the parents will be watching the other drunk Santa, downing eggnog. They'll be too distracted to notice a couple of strangers doing laps around the Christmas tree, filling their sacks to the brim." Bill tapped his chin, pondering the plan. "I still think we should just go get a few lumps of coal." Molly smirked. "Oh come on, they've got to have some good loot there - video games, the latest gadgets. Trust me, we'll be knocking on the doors of the naughty list this Christmas!"
id
"gen-wz0szgp4IiPk7ZZArdyzXibImz0j"
total_cost
0.0013625
upstream_id
"5f8e9ca1LZrK346M"
created_at
"2024-07-19T15:34:49.402532+00:00"
Friday, July 19th, 2024 at 3:34:49 PM GMT+00:00
model
"alpindale/magnum-72b"
app_id
182717
streamed
false
cancelled
false
provider_name
"Mancer"
latency
15695
moderation_latency
(null)
generation_time
0
finish_reason
"stop"
tokens_prompt
67
tokens_completion
242
native_tokens_prompt
72
native_tokens_completion
243
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.0013625
Evaluation details
Result
Evaluator
Details
Meta Data
1.0360%
Matches word count
n/a
neededClean
false
words
174
99.8169%
Dialogue to Total Word Ratio
Ratio: 87.93%, Deviation: 2.07%
neededClean
false
wordsTotal
174
wordsDialogue
153
50.4265%