NC Bench
Overview
About
Tests
Models
Model
Llama 3.2 11B (Vision)
Test
Dialogue tags
Scenario
Write 200 words with 10% dialogue
Run ID
2024-09-26-01J8PZ22ZDB121YF1MSMXDY4HS
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
The conference room was filled with the faint scent of stale donuts and broken dreams. Molly and Bill sat across from each other, sipping lukewarm coffee from matching "World's Okayest Thief" mugs. Molly fanned out a deck of glossy party invitations on the table, scrutinizing the layout of each one. "Santa's Workshop is too secure, they've got motion detectors the size of small cars. Mrs. Claus's Cookies might be an option, but I heard their security team is staffed by retired hitmen with a vendetta against toys." Bill snorted, "Those guys are old news. I've got a better idea. Kidzilla's Karaoke Palace is hiring for a DJ. I managed to 'accidentally' slip a uniform my way. I'll blend in as the emcee, distract the parents while you scope out the gift table." Molly's eyes widened as she pored over the invitations once more. "Kidzilla's has an open bar for grown-ups and a petting zoo. The odds of getting out of there undetected are higher than getting a free donut at this pathetic conference room."
id
"gen-SgHidaobBpfYlPgECB9sLj1wSu4R"
total_cost
0.000049734
upstream_id
"chat-eda1a47c97c841269b20f41ba593172b"
created_at
"2024-09-26T10:47:06.831018+00:00"
Thursday, September 26th, 2024 at 10:47:06 AM GMT+00:00
model
"meta-llama/llama-3.2-11b-vision-instruct"
app_id
182717
streamed
true
cancelled
false
provider_name
"Together"
latency
135
moderation_latency
(null)
generation_time
1431
finish_reason
"stop"
tokens_prompt
68
tokens_completion
235
native_tokens_prompt
71
native_tokens_completion
236
native_tokens_reasoning
(null)
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.000049734
Evaluation details
Result
Evaluator
Details
Meta Data
2.0116%
Matches word count
n/a
neededClean
false
words
175
0.0000%
Dialogue to Total Word Ratio
Ratio: 65.71%, Deviation: 55.71%
neededClean
false
wordsTotal
175
wordsDialogue
115
1.0058%