NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1K55391GN00EGV40G326KTM
Run Details
Model:
Muse Spark 1.3 (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
1m 4s
Cost:
$0.031781
Provider:
Meta
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.29%
Dialogue to Total Word Ratio
(details)
99.6446%
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
The van smelled like old fries and bad decisions as Molly spread the crayon-drawn map across the dashboard. Rain hammered the cracked windshield while a half-deflated unicorn balloon bobbed from the rearview mirror and the radio played cheerful birthday jingles on loop. Bill picked nervous glitter from his black hoodie and eyed the parking lot full of minivans, bounce houses, and exhausted parents wrangling sugar-fueled toddlers with juice boxes. Molly grinned, tapping her clipboard, already counting wrapping paper, ribbons, and glorious untagged birthday loot waiting inside. "Okay Bill, the target is Timmy Johnson's seventh birthday party at Chuckle Cheese Palace this Saturday at noon," Molly said. "Chuckle Cheese Palace? Molly, that's sacred ground. I had my eighth birthday there. I still have trauma from the ball pit," Bill said. "Focus. Twenty kids, twenty presents, one distracted clown making balloon animals. It's the perfect score," Molly said. "Stealing presents from kids feels low, even for us. Can't we rob a bank like normal criminals?" Bill asked. "Banks have cameras and vaults and guards with actual guns. Timmy's mom has a Pinterest board and a frazzled babysitter. Which sounds easier?" Molly asked. "Fair point. But what's the plan? We can't just walk in and grab Paw Patrol gift bags," Bill said. "We go undercover. I'll be Princess Sprinkles the face painter. You be Captain Giggles the clown," Molly said. "I am not wearing clown shoes again. Last time I got athlete's foot and emotional damage," Bill said. "You'll wear the shoes, Bill. The giant table of unopened presents is right by the cake station, totally unguarded during musical chairs," Molly said. "Musical chairs? That's when we strike? While innocent children fight over chairs to bad pop music?" Bill asked. "Exactly. Phase one, you distract them with your amazing balloon poodle. Phase two, I wheel the gift cart to the emergency exit," Molly said. "I still feel guilty. What if Timmy cries? What if little Susie with the pigtails gives me those big sad eyes?" Bill asked. "Bill, we are professionals. We do not get emotionally compromised by pigtails. Think about the haul. Legos, video games, gift cards," Molly said. "Okay, okay. But I want the big box with the dinosaur on it. I always wanted a Roar Rex Turbo Dino," Bill said. "No. Everything goes to the fence first, then we split sixty forty since it's my plan and my van and my fake mustache collection," Molly said. "Sixty forty? That's outrageous. I'm risking clown humiliation here. Fifty fifty or I tell Timmy's mom," Bill said. "Fine, fifty fifty. But you have to wrap the getaway car in birthday streamers so we blend in," Molly said. "Deal. And if we get caught, we're just two very lost adults who really love cake and pinatas," Bill said. "Exactly. No one suspects adults crying at a magic show. It's camouflage," Molly said. Molly folded the map as the neon sign of Chuckle Cheese Palace blinked in the distance like a tacky beacon of opportunity.
created_at
"2026-09-03T08:11:21.255Z"
9/3/2026, 8:11:21 AM
model
"meta/muse-spark-1.3-20260902"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
61741
moderation_latency
(null)
generation_time
63608
tokens_prompt
75
tokens_completion
760
native_tokens_prompt
68
native_tokens_completion
7458
native_tokens_completion_images
0
native_tokens_reasoning
6765
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
0.0317815
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788423081-Y02X68OLdezpTAA8g6Rh"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788423081-yVfWHBIhUHPaeSbNfLJI"
upstream_id
"resp_6a992bac7a0d27f7e1ba431e"
provider_responses
0
endpoint_id
"cf4f1b4e-1719-4b65-9111-7dd7635e5a2f"
id
"resp_6a992bac7a0d27f7e1ba431e"
is_byok
false
latency
3178
model_permaslug
"meta/muse-spark-1.3-20260902"
provider_name
"Meta"
status
200
total_cost
0.0317815
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
99.29%
Dialogue to Total Word Ratio
Ratio: 70.38%, Deviation: 0.38%
neededClean
false
wordsTotal
503
wordsDialogue
354
99.6446%