Model

Llama 3.2 90B (Vision)

Test

Dialogue tags

Scenario

Write 200 words with 10% dialogue

Run ID

2024-09-26-01J8Q0NWX5WD2N2DX9J7DSYGC4

Model input

User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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
The stimulation of the alarm clock sounded through the cluttered room, dust covering every inch, proof of Molly and Bill's ongoing preparation for their next heist. They stood at their makeshift planning table, a thorough map of the suburban children's party they planned to raid laid out before them. Party hats and balloons surrounded the area of the supposed cake-filled pagoda. Their attire was anything but ordinary – Molly sported a trench coat, albeit a worn-out one, whereas Bill was in tattered grease-stained overalls. "Easiest heist of the century," Molly claimed with excitement. Bill lifted his gaze from the map and grinned. Lifting a grease-stained thumb, he started tracing the fence's curve. A droplet of clearer blue was the assembly place near the exit. Bill instructed Molly on cover and escape strategies alone. Blue - indicated the positions of watching parents needed careful evading strategies "Easy in and out," Bill added smoothly. Careful gazes weaved back and forth studying maps laid bare to any probabilities raising quickly accessible test the manner of same outlines off fully attractive presentation of seeing thrown festivities today before she innerren's blimp augment fiery fer insurer-broken heroes kept easy dots production mode. The scene confirmed truth if adul harmless kids ostensibly kept age humanity amazingly tough theor prevent explanation only.
id"gen-OJeTFiGDyD0ZArQuvMeG2CwDP47G" 
total_cost0.0001456
upstream_id"chat-6220f6282e9844d799d9698b8feff81d" 
created_at"2024-09-26T11:15:33.294153+00:00" Thursday, September 26th, 2024 at 11:15:33 AM GMT+00:00
model"meta-llama/llama-3.2-90b-vision-instruct" 
app_id182717
streamedtrue
cancelledfalse
provider_name"Hyperbolic" 
latency181
moderation_latency(null)
generation_time10233
finish_reason"stop" 
tokens_prompt68
tokens_completion266
native_tokens_prompt96
native_tokens_completion268
native_tokens_reasoning(null)
num_media_prompt(null)
num_media_completion(null)
origin"https://ncbench.com/" 
usage0.0001456

Evaluation details

Result Evaluator Details Meta Data
35.0022% Matches word count n/a
neededCleanfalse
words218
88.7937% Dialogue to Total Word Ratio
Ratio: 4.13%, Deviation: 5.87%
neededCleanfalse
wordsTotal218
wordsDialogue9
61.8980%