NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1K5WFF1A6GNZ7VBBV8D0X60
Run Details
Model:
Muse Spark 1.2 (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
22.4s
Cost:
$0.029040
Provider:
Meta
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.87%
Dialogue to Total Word Ratio
(details)
99.9371%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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 crouched behind the garden hedge, binoculars trained on the bouncing castle wheezing in the suburban backyard. Bill lay beside her, noisily slurping a juice box he had already stolen from the recycling bin. A large banner read Happy Birthday Timothy in letters that had started peeling. The gift table sagged beneath wrapped boxes, ribbons shimmering in the heat. A mother in sunglasses patrolled the perimeter with a clipboard, ticking off names as children shrieked inside the inflatable fortress. "Okay, remind me again why we are robbing a six-year-old's birthday party?" Bill asked. "Because Timothy's mom went overboard," Molly whispered. "There are at least twenty presents stacked by the gift table. That's a haul, Bill. That's college fund money." "That's also stealing from children, Molly. Even for us that's low," Bill said. "We're not stealing from children, we're liberating plastic toys from ungrateful children," Molly said. "Do you know what kids do with presents? They play with the box." "Fair point, my nephew still prefers the remote control," Bill said. "The plan is simple," Molly said. "At three o'clock they bring out the cake. Every adult will be singing and every kid will be screaming. That's our window." "How long is the window?" Bill asked. "Approximately the length of Happy Birthday sung aggressively off-key by twenty parents," Molly said. "So ninety seconds," Bill said. "Exactly. You will dress as the clown," Molly said. Bill adjusted his fake red nose, which kept slipping down his sweaty face. "I am not dressing as a clown, Molly. I have boundaries," Bill said. "You literally wore a nun costume for the museum job," Molly said. "That was a sexy nun costume, completely different," Bill said. "This is a children's clown costume, it smells faintly of old face paint and regret," Molly said. "You will distract them with balloon animals while I grab the presents." "What if I can't make balloon animals?" Bill asked. "Then make balloon tragedies, just keep them looking," Molly said. "And how are you getting the presents out?" Bill asked. "I brought the wagon," Molly whispered. "The little red one with the squeaky wheel." "The squeaky wheel is not stealthy, Molly," Bill said. "I oiled it with juice box residue," Molly said. "It will be silent as a whisper." "There's a mom guarding the presents like a dragon," Bill said. "Look, the one with the clipboard." "That's why I brought sleeping powder," Molly said. "You mean sugar?" Bill asked. "No, actual sleeping powder, I bought it from a magician," Molly said. "You got scammed by a magician, Molly," Bill said. "Just watch me work," Molly said. "You do the clown thing, I'll do the old gift-nap." "What do we do if a kid cries?" Bill asked. "We panic and run, same as always," Molly said. "Do we at least get cake after?" Bill asked. "If we don't get caught, you can have all the cake you want," Molly said. "If we do get caught, we're explaining clown theft to the police."
created_at
"2026-09-03T08:24:07.399Z"
9/3/2026, 8:24:07 AM
model
"meta/muse-spark-1.2-20260805"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2604
moderation_latency
(null)
generation_time
22381
tokens_prompt
75
tokens_completion
784
native_tokens_prompt
68
native_tokens_completion
6813
native_tokens_completion_images
0
native_tokens_reasoning
6110
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"completed"
service_tier
"auto"
usage
0.02904025
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788423847-O1BCipIdrJelsHnNYCMP"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788423847-uHLkMC1b8EQkqDrJjmLP"
upstream_id
"resp_6a992ea7e2cdb4701dce4aee"
provider_responses
0
endpoint_id
"ffc8e76e-808a-44d2-bd21-c6ac16e721cb"
id
"resp_6a992ea7e2cdb4701dce4aee"
is_byok
false
latency
840
model_permaslug
"meta/muse-spark-1.2-20260805"
provider_name
"Meta"
status
200
total_cost
0.02904025
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Meta"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
500
99.87%
Dialogue to Total Word Ratio
Ratio: 69.84%, Deviation: 0.16%
neededClean
false
wordsTotal
504
wordsDialogue
352
99.9371%