Model

Llama 3.2 90B (Vision)

Test

Dialogue tags

Scenario

Write 200 words with 90% dialogue

Run ID

2024-09-26-01J8Q0RM19W0PF9KYJMT22BAYM

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

AI
"I'm telling you, Bill, it's genius," Molly said, a mischievous glint in her eye. "A children's birthday party is the perfect mark. Nobody expects to get robbed at a bouncy castle." "Except all the parents with anxiety disorders," Bill replied dryly. "They're probably expecting a vodka theft ring." Molly waved her hand dismissively. "Details, details. The point is, we get in, grab the presents, and get out before anyone even notices the piƱata's been breached." "What's the plan if we get caught?" Bill asked, inspecting the lock on the cardboard box filling the trunk of the car. "Oh, we'll just pretend to be the entertainment," Molly said, producing a pair of oversized, novelty sunglasses from her bag. "I've got it all figured out. I've been practicing my best 'accidental party crasher' face." Bill raised an eyebrow. "That's your entire plan?" Molly grinned. "Hey, I'm a master of improv. I once convinced a downtown ATM that I was a police officer." Bill shook his head. "You were wearing a bathrobe at the time, weren't you?" Molly shrugged. "Details."
id"gen-8V2AbSwSKUZyKFnpxIo2ioz0CJLo" 
total_cost0.000351
upstream_id"chat-739dc11f915d4126a7919acd5c26b03d" 
created_at"2024-09-26T11:16:57.572384+00:00" Thursday, September 26th, 2024 at 11:16:57 AM GMT+00:00
model"meta-llama/llama-3.2-90b-vision-instruct" 
app_id182717
streamedtrue
cancelledfalse
provider_name"Together" 
latency204
moderation_latency(null)
generation_time5205
finish_reason"stop" 
tokens_prompt68
tokens_completion253
native_tokens_prompt71
native_tokens_completion254
native_tokens_reasoning(null)
num_media_prompt(null)
num_media_completion(null)
origin"https://ncbench.com/" 
usage0.000351

Evaluation details

Result Evaluator Details Meta Data
6.0907% Matches word count n/a
neededCleanfalse
words177
0.0000% Dialogue to Total Word Ratio
Ratio: 68.93%, Deviation: 21.07%
neededCleanfalse
wordsTotal177
wordsDialogue122
3.0453%