NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2024-07-23-01J3G80W4ZP7Y73HNKN6005TNW
Run Details
Model:
Llama 3.1 8B
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
1.3s
Cost:
$0.000066
Provider:
Fireworks
Evaluation summary
Score
Evaluator
Details Link
43.38%
Matches word count
(details)
63.61%
Dialogue to Total Word Ratio
(details)
53.4956%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 50% 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
Molly stood in the cramped, dimly lit storage room, surrounded by stacks of dusty boxes and cobweb-covered awards. Bill scribbled notes on a crumpled napkin, a grin spreading across his face. "Alright, so we hit the party Saturday at two pm, when everyone's too busy stuffing their faces with cake to notice us snooping around," Bill said, never looking up. "And when I give the signal, you make your way to the present pile. I'll take care of the adults, snag the iPads and credit cards," Molly replied, her eyes scanning the plan as if searching for any loopholes. "Got it. But what about the clown? Apparently, he's been training in hand-to-hand combat techniques specifically designed for deflecting awkward birthday hugs," Bill warned, his brow furrowed in concern. Molly rolled her eyes, a dry smile spreading across her face. "Leave the clown to me. I've been practicing my 'I'm-a-party-photographer' disguise. I'll make sure he's too distracted to notice us swiping the good stuff." Bill snorted, nodded, and tucked the napkin into his back pocket. "Time to write our next bestselling 'The Getaway' novel."
id
"gen-Ey04glrknH0PUiuZJPvIoZvWTIIB"
total_cost
0.0000656
upstream_id
"fdf7cff8-78d5-42e3-86e9-cab224365e6c"
created_at
"2024-07-23T16:50:33.821627+00:00"
7/23/2024, 4:50:33 PM
model
"meta-llama/llama-3.1-8b-instruct"
app_id
182717
streamed
false
cancelled
false
provider_name
"Fireworks"
latency
1177
moderation_latency
(null)
generation_time
0
finish_reason
"stop"
tokens_prompt
102
tokens_completion
256
native_tokens_prompt
70
native_tokens_completion
258
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.0000656
Evaluation details
Result
Evaluator
Details
Meta Data
43.38%
Matches word count
n/a
neededClean
false
words
183
63.61%
Dialogue to Total Word Ratio
Ratio: 58.20%, Deviation: 8.20%
neededClean
false
wordsTotal
189
wordsDialogue
110
53.4956%