NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1K5MV4NWE7PPHGC9XNV016W
Run Details
Model:
Muse Spark 1.3 (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
2m 3s
Cost:
$0.039300
Provider:
Meta
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
100.0000%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 cracked party invitation across the hood of their dented van and smoothed it flat with a greasy sleeve. Rain drummed on the roof and made the backyard map bleed pink ink. "We are really doing this, Bill? Robbing a tiny birthday?" Bill looked up from polishing a stolen traffic cone and grinned like this was a bank vault. "It's not robbing, Molly. It's aggressive present redistribution." Molly groaned and pointed at the handwritten details with a chewed pen. "There are twenty kids, one clown, and a mom with eyes like a hawk." Bill tapped the clown listed on the entertainment schedule and winked with total confidence. "Perfect. Nobody suspects the extra clown." Molly stared at the rainbow clown suit leaking from their duffel bag and shuddered. "I am not wearing the clown suit again. It squeaks when I walk." Bill folded the squeaky suit away and offered her a stained cardigan instead with ceremony. "Fine. You're the tired mom with yoga pants and cold coffee." Molly sniffed the cardigan, found old goldfish crackers in the pocket, and nodded slowly. "That I can do. What's your brilliant entrance plan, genius?" Bill unrolled a crude crayon diagram of the living room with arrows, exits, and a gift table marked with stars. "We walk in with a big wrapped box and look exhausted." Molly studied the diagram, tapped the back door near the kitchen, and frowned at the narrow fence. "And the getaway? We can't outrun crazy toddlers." Bill traced an escape route past the bounce house with a french fry and looked proud. "We park the van by the bounce house and float away casual." Molly imagined sprinting with armfuls of glittery boxes while toddlers screamed for frosting. "You mean stumble away while carrying seventeen Paw Patrol sets?" Bill laughed and patted the empty van where the seats had been replaced by blankets. "Exactly. Nobody chases a mom carrying gifts. It's camouflage." Molly checked the driveway photo on her phone and counted three minivans blocking easy access. "If that clown touches me, I'm stealing his balloon animals too." Bill promised to distract the performer with endless balloon requests and fake enthusiasm. "Deal. You grab presents, I secure cake. For morale." "You are the worst. Let's steal birthday joy." The van sputtered to life as Molly braided her hair into a frazzled mom bun and smeared grape jelly on her cardigan for authenticity. Bill squeezed into an oversized Happy Birthday shirt that read Best Dad and practiced his exhausted wave in the cracked mirror. They loaded one enormous empty box wrapped in newspaper comics and labeled it From Aunt Linda. Outside the suburban house glowed with string lights while shrieks and pop music leaked through open windows. Molly gripped the door handle and exhaled. This was ridiculous, petty, and strangely thrilling. Bill honked twice, grabbed the frosting stained map, and whispered that history favors the bold and the mildly embarrassing. They waddled toward glitter doom with confidence only amateurs possess.
created_at
"2026-09-03T08:19:57.212Z"
9/3/2026, 8:19:57 AM
model
"meta/muse-spark-1.3-20260902"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
120003
moderation_latency
(null)
generation_time
123231
tokens_prompt
75
tokens_completion
778
native_tokens_prompt
68
native_tokens_completion
9227
native_tokens_completion_images
0
native_tokens_reasoning
8577
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
0.03929975
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788423597-dridT9ymUNJqdrijXhDT"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788423597-vFdvU0UzxlGlSJxsNuPg"
upstream_id
"resp_6a992db0f989517439f145fc"
provider_responses
0
endpoint_id
"cf4f1b4e-1719-4b65-9111-7dd7635e5a2f"
id
"resp_6a992db0f989517439f145fc"
is_byok
false
latency
3222
model_permaslug
"meta/muse-spark-1.3-20260902"
provider_name
"Meta"
status
200
total_cost
0.03929975
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
100.00%
Dialogue to Total Word Ratio
Ratio: 30.00%, Deviation: 0.00%
neededClean
false
wordsTotal
500
wordsDialogue
150
100.0000%