NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1K5ENQPXYJC9NY0Y37QDFJ3
Run Details
Model:
Muse Spark 1.3 (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
1m 36s
Cost:
$0.038046
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 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 minivan with binoculars shaped like juice boxes. Bill ate an orange slice while studying the suburban house across the street. Balloons bobbed over the fence and a banner screamed HAPPY 6TH BIRTHDAY LEO in glitter. Molly whispered, "Okay, target is loaded with presents. I count fourteen boxes, two bikes, and one suspiciously large LEGO castle." Bill nodded, "Fourteen boxes? That's amateur hour. Last Christmas we cleared twenty houses before cocoa." He adjusted his clown nose disguise, which squeaked with guilt. Molly snorted, "You look like a sad tomato. The kids will love you, which is perfect for our cover." Bill grinned, "Right, Operation Birthday Grab. I walk in as Clown Billo, you pose as Magical Assistant Sparkles." Molly twirled a wand that shot glitter and mild electroshocks. Molly said, "I distract them with terrible magic tricks while you stuff presents into the laundry cart of doom." Bill asked, "What if they recognize terrible magic? Last time you pulled a rabbit out and it filed for workers comp." The clown across the street popped a balloon animal and three toddlers screamed with delight. Molly hissed, "Focus, Bill. We go over the side fence at cake time. Sugar crash equals low security." Bill whispered, "Cake time is genius. Nobody guards presents when chocolate frosting is legally available." He pulled a blueprint drawn in crayon from his squeaky shoe. Bill explained, "Front door is clown entrance. Back door is cake exit. Laundry cart waits by the trampoline for extraction." Molly laughed, "You drew laser sharks guarding the bounce house. Are those real or budget fears?" Bill muttered, "Both. Never underestimate six year olds hopped up on juice. They bite like tiny raccoons." A mom wheeled out a towering cake shaped like a dinosaur eating a pirate ship. Molly gasped, "Oh no, cake now! Go go go! Remember, act magical, steal fast, tip the clown." Bill squeaked, "I was born ready! If caught, we claim we are extreme party planners collecting gifts for quality inspection." They waddled toward the fence pushing the laundry cart covered in birthday stickers. Molly ordered, "Keep squeaking that nose! Louder! Annoying equals authentic clown behavior in the wild." Bill yelled, "Birthday heist is a go! I see frosting, I see chaos, I see at least three juice pouches I can bribe witnesses with right now! Wish me greasy luck please!" Molly replied, "Remember the code! If the clown cries, abort! If the mom offers you cake, accept and run! If Leo asks for magic, vanish with his scooter collection immediately! No heroic balloon fights!" Bill shouted, "Deal! You pull the rabbit, I pull the presents! We will be legends! They will write songs about the Great PiƱata Caper and our daring frosting escape tonight! Bring extra tote bags quickly!" Molly shouted back, "Always! Steal every truck, doll, and glittery unicorn! Leave no bow untaken! Just save me some cake because crime burns calories and I am starving already today! Meet at minivan soonest! Go wild!"
created_at
"2026-09-03T08:16:35.068Z"
9/3/2026, 8:16:35 AM
model
"meta/muse-spark-1.3-20260902"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
91643
moderation_latency
(null)
generation_time
95823
tokens_prompt
75
tokens_completion
759
native_tokens_prompt
68
native_tokens_completion
8932
native_tokens_completion_images
0
native_tokens_reasoning
8235
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
0.038046
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788423395-ppMxQdPctO6di6ISJ77L"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788423395-uCpGVRCDSmoaPWJAifqP"
upstream_id
"resp_6a992ce377dc2cf27897431d"
provider_responses
0
endpoint_id
"cf4f1b4e-1719-4b65-9111-7dd7635e5a2f"
id
"resp_6a992ce377dc2cf27897431d"
is_byok
false
latency
750
model_permaslug
"meta/muse-spark-1.3-20260902"
provider_name
"Meta"
status
200
total_cost
0.038046
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: 70.00%, Deviation: 0.00%
neededClean
false
wordsTotal
500
wordsDialogue
350
100.0000%