NC Bench
Overview
About
Tests
Models
Model
Llama 3.1 Nemotron 70B
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-10-16-01JABHZWX44AZ69VDD40CKZFAX
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'm telling you, Bill, this is the heist of the century," Molly said, eyes shining with excitement. "Oh please, it's a kids' birthday party. The security is going to be, what, four frazzled parents and a bouncy castle?" Bill raised an eyebrow. "Exactly! That's what makes it genius. No one expects the presents to get jacked at a five-year-old's shindig. We'll be in and out before they can say 'piñata'." "Ugh, don't remind me. I still have PTSD from that last job where we had to escape through the ball pit." "Aw, come on, it was fun! You got a good story out of it. 'The Great Ball Pit Breakout' – I'm still waiting for you to write that memoir." "Ha ha, very funny, Molly. So, what's the plan for today? You've got the party layout, I assume?" "Of course. I've been casing this joint for weeks. Okay, so the presents are on a table near the cake... we create a diversion, grab the loot, and make a run for the Venezian ice cream truck parked outside." "Sold. But if I get stuck in a bounce house again, I'm never working with you again."
id
"gen-1729112306-vrMMvNdCp9Q2paZswfzP"
upstream_id
"cmpl-Rzy9g7w2Oi3Lj58S2I74KdZP"
total_cost
0.00013285
cache_discount
(null)
created_at
"2024-10-16T20:58:38.902577+00:00"
Wednesday, October 16th, 2024 at 8:58:38 PM GMT+00:00
model
"nvidia/llama-3.1-nemotron-70b-instruct"
app_id
182717
streamed
true
cancelled
false
provider_name
"DeepInfra"
latency
68
moderation_latency
(null)
generation_time
11688
finish_reason
"stop"
tokens_prompt
102
tokens_completion
270
native_tokens_prompt
71
native_tokens_completion
270
native_tokens_reasoning
(null)
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.00013285
Evaluation details
Result
Evaluator
Details
Meta Data
99.3769%
Matches word count
n/a
neededClean
false
words
195
94.5225%
Dialogue to Total Word Ratio
Ratio: 94.87%, Deviation: 4.87%
neededClean
false
wordsTotal
195
wordsDialogue
185
96.9497%