NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2024-10-16-01JABJ04BHDDBMWVNDDNWZYTP2
Run Details
Model:
Llama 3.1 Nemotron 70B
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
9.3s
Cost:
$0.000126
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.08%
Matches word count
(details)
99.94%
Dialogue to Total Word Ratio
(details)
50.0127%
Overall
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, it's a goldmine. A dozen sugar-high kids, distracted by a guy in a sweaty costume, and a pile of unguarded gifts. It's like they're begging to be robbed." Molly said, eyes shining with mischief. "Unethical, Molly. We're talking about kids' birthday presents here. What's next, swiping from the Salvation Army?" Bill countered, rubbing the back of his neck. "Oh, come on! These aren't exactly disadvantaged youth. The theme is 'Princess & Pirate Luxury Bash' – these kids are born with silver spoons. We're just... redistributing the wealth." Molly reasoned. "Arguably, to our own, very selfish pockets. And what about the security? I heard the party planner hired 'Bouncer Bob' – the guy's a former cop." "Bouncer Bob or not, I've got a plan. We'll blend in with the face-painting station. Who suspects a woman with a glitter mustache and a man in a 'World's Okayest Uncle' t-shirt?" Molly deadpanned. Bill sighed, "You know I'm in, but if I get stomped by a toddler in a tiara, I'm blaming you."
id
"gen-1729112314-qVpy4rnZD5KK1ZHSQh5a"
upstream_id
"cmpl-RlHdi3rofb908TLWbvgn7yHr"
total_cost
0.00012645
cache_discount
(null)
created_at
"2024-10-16T20:58:44.012934+00:00"
10/16/2024, 8:58:44 PM
model
"nvidia/llama-3.1-nemotron-70b-instruct"
app_id
182717
streamed
true
cancelled
false
provider_name
"DeepInfra"
latency
92
moderation_latency
(null)
generation_time
9023
finish_reason
"stop"
tokens_prompt
102
tokens_completion
254
native_tokens_prompt
71
native_tokens_completion
254
native_tokens_reasoning
(null)
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.00012645
Evaluation details
Result
Evaluator
Details
Meta Data
0.08%
Matches word count
n/a
neededClean
false
words
171
99.94%
Dialogue to Total Word Ratio
Ratio: 88.44%, Deviation: 1.56%
neededClean
false
wordsTotal
173
wordsDialogue
153
50.0127%