NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2024-07-23-01J3G809PJ3DAMHB28116KFSEW
Run Details
Model:
Llama 3.1 70B
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
4.1s
Cost:
$0.000322
Provider:
Fireworks
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
68.3935%
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, this is the score of a lifetime. A children's party, full of unsuspecting parents and kids so distracted by the bouncy castle they won't even know what's happening." "A bouncy castle? You're counting on a bouncy castle as our getaway distraction?" "Hey, it's a classic move. Works every time. Besides, have you seen the gift list for this thing? iPads, Xboxes, the latest Lego sets... we'll be set for months if we can pull this off." "I don't know, Molly. I've got a bad feeling about this. What if we get caught? I don't fancy doing time for nicking a toddler's Play-Doh." "Play-Doh? You're not thinking about the big picture here. We're not in this for the craft supplies. Although, if we do happen to get stuck with a few boxes, I hear it's great for stress relief." "Stress relief? You're the one who's always getting us into these situations. I'm starting to think you're the one who needs the stress relief." "Oh, come on. It'll be fun, I promise. And who doesn't love a good piñata?" "Exactly. Who doesn't love a good piñata? That's exactly what I'm worried about. Witnesses, Molly. Lots of tiny, screaming witnesses."
id
"gen-9x3HAmh65n5x8QXWBOhaQ5dD8QAl"
total_cost
0.0003222
upstream_id
"c92b618b-0ba3-4156-b717-5afa4eb5db2f"
created_at
"2024-07-23T16:50:17.764866+00:00"
7/23/2024, 4:50:17 PM
model
"meta-llama/llama-3.1-70b-instruct"
app_id
182717
streamed
false
cancelled
false
provider_name
"Fireworks"
latency
3500
moderation_latency
(null)
generation_time
280
finish_reason
"stop"
tokens_prompt
102
tokens_completion
286
native_tokens_prompt
70
native_tokens_completion
288
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.0003222
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
201
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
203
wordsDialogue
203
68.3935%