NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1K4QJF0V7YXAJDC19P9BXE8
Run Details
Model:
Muse Spark 1.2 (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
26.8s
Cost:
$0.024633
Provider:
Meta
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
98.28%
Dialogue to Total Word Ratio
(details)
99.1424%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 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 spread the blueprint across the stolen pizza box, smoothing creases with the flat of her hand. The paper showed the Johnson house, complete with bouncy castle and a fortified living room labeled PRESENT MOUNTAIN. Bill leaned over her shoulder, his fake mustache already peeling in the summer heat. "So the target is a six-year-old's birthday party?" Molly tapped a red X over the backyard. "Not just any party, Bill. Timmy Johnson is turning six and his grandmother always gives savings bonds." Bill whistled low and adjusted his clown nose, which squeaked mournfully. "You want us to infiltrate a children's party dressed as entertainers to steal Bluey-themed gift wrap?" She pulled two duffel bags from under the table, unzipping one to reveal a tangled mass of balloons. "Exactly. We walk in, we do ten minutes of terrible magic tricks, we walk out with a sack of presents bigger than Santa's." Bill stared at the bag, his confidence visibly deflating like one of the balloons. "Molly, I didn't become a professional thief to steal juice boxes and LEGO sets." Molly fixed him with a stare that had once made a bank guard hand over his own lunch. "Think bigger, Bill. Last year Timmy got a gaming console, a tablet, and a solid gold piggy bank from his uncle." She flipped the blueprint over, revealing a hand-drawn timeline marked with cake cutting and nap time. "What's the plan for the parents? There will be like twenty moms filming everything on their phones." Molly grinned and produced a glitter-covered invitation that read Admit Two Entertainers. "We are the entertainment. You do balloon animals, I do face painting, and at three o'clock we cause a distraction." Bill picked up a limp balloon and tried to twist it, immediately popping it with a loud bang. "What kind of distraction works on a sugar-crazed horde of six-year-olds?" "You will announce that the bouncy castle is free forever and open to everyone at once." Bill imagined the stampede, the screaming, the absolute chaos of thirty children bouncing into oblivion. "That's evil, Molly. That's genuinely evil. I love it. When do we grab the loot?" Molly pointed to the window on the blueprint, the one beside the present table with the faulty lock. "While they are bouncing, we slide the presents out the window into the van. Easy and quiet." Bill nodded slowly, finally seeing the genius of targeting an audience too short to see over the table. "And if a kid catches us? What do we say to a crying six-year-old with frosting on his face?" "We tell him we are taking the presents to the North Pole for quality control testing. He will believe us." Outside their hideout, a distant sound of children laughing drifted through the alley, unaware of the impending heist. Bill sighed, pulled his drooping mustache straight, and finally reached for the face paint with trembling determination. "Fine, but if I have to paint a unicorn, you are carrying the box of LEGOs."
created_at
"2026-09-03T08:03:58.054Z"
9/3/2026, 8:03:58 AM
model
"meta/muse-spark-1.2-20260805"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2487
moderation_latency
(null)
generation_time
26726
tokens_prompt
75
tokens_completion
773
native_tokens_prompt
68
native_tokens_completion
5776
native_tokens_completion_images
0
native_tokens_reasoning
5100
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
0.024633
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788422638-zaWYkfgam3ynu5Lhq9Dv"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788422638-6hb1Q0SWCGVkDnyKB0lt"
upstream_id
"resp_6a9929ee0fafe57b903646b7"
provider_responses
0
endpoint_id
"ffc8e76e-808a-44d2-bd21-c6ac16e721cb"
id
"resp_6a9929ee0fafe57b903646b7"
is_byok
false
latency
391
model_permaslug
"meta/muse-spark-1.2-20260805"
provider_name
"Meta"
status
200
total_cost
0.024633
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
98.28%
Dialogue to Total Word Ratio
Ratio: 50.59%, Deviation: 0.59%
neededClean
false
wordsTotal
510
wordsDialogue
258
99.1424%