NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYD7FK5J7HYW4F3KVHHVQGD1
Run Details
Model:
Muse Spark 1.1 (Reasoning, Minimal)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
15.0s
Cost:
$0.008042
Provider:
Meta
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
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
"Okay, walk me through it again, but slow this time because my clown shoes keep falling off," Molly said. "They're not clown shoes, Molly, they're just your regular shoes two sizes too big because you stole them off the actual clown," Bill said. "I borrowed them. With aggressive intent. Focus. What's the plan?" "The plan is we go to Timmy's seventh birthday party, we act natural, and we steal every single present on that table." "Define natural. Because you don't look natural around children. Last time you saw a bouncy house you tried to rob it." "I did not try to rob a bouncy house. I got stuck in a bouncy house and then I panicked and demanded everyone's wallets. Two very different things. Look, it's simple. We get invited as entertainers." "We are not entertainers. You cried during musical chairs." "Okay, fine, we are magicians. We say abracadabra, we make the presents disappear. Comedic gold." "Bill, you are the worst magician I have ever met. You once tried to pull a rabbit out of a hat and pulled out a subpoena." "That rabbit owed me money. Listen, the backyard gate will be open. The mom is distracted. The dad is trying to assemble a trampoline with a butter knife. We walk in, we grab the tablecloth like a cartoon - fwip! - all the presents in one sack." "And then what? We sprint down the street with a sack labeled SUPERHERO BIRTHDAY BOY while eight seven-year-olds chase us screaming?" "Exactly! No one will suspect two adults!" "Everyone will suspect two adults. Especially two adults carrying forty pounds of Paw Patrol wrapping paper." "So we need a distraction. That's where you come in. You do your balloon animals." "I don't know how to do balloon animals. Last time I tried I made something that looked like a sad intestine and made a child call his therapist." "Perfect! You traumatize them with the balloons, they all look at you, I grab the loot. We meet by the van. The van that does not say FREE CANDY anymore since I painted over it." "You painted over it with STILL FREE CANDY BUT IRONICALLY. That is worse. Why are we even stealing children's presents anyway? There's nothing good in there. It's all slime and gift cards for five dollars." "One of them is a limited edition Robo-Dino Ultra. Timmy's rich grandma bought it. It's worth eight hundred bucks on eBay. Also, principle." "What principle? Steal from the rich and give to ourselves because we are bad at jobs?" "The principle that no one ever invited us to these parties as kids, Molly! We are reclaiming our childhood!" "We were not invited because we tried to steal the presents! Bill, the operating budget for this heist is sixty bucks in gas and a clown nose that smells like cheese." "So it's a low-overhead operation. That's good business." "It's a terrible business. Okay, new plan. We go in, we act like parents. You put on that mustache." "This mustache makes me look distinguished. Like a dad who knows about taxes." "That mustache makes you look like you lost a fight with a Sharpie. Fine. I will go in first. I will say, Wow, Timmy, what a great pile of totally secure presents no one is watching. Then you grab." "And if someone asks who we are?" "We say we're Timmy's uncles. Uncle Molly and Uncle Bill. Totally believable." "Timmy is an only child and his parents are both only children. He has no uncles." "Then we're... magician uncles. From... the magic union." "You can't just say union and make it legit." "Watch me. I have a clipboard. Clipboards make everything legit." "I swear if we get beaten up by a bunch of second graders again I'm leaving you there." "They did not beat us up last time. It was a tactical retreat from children with juice boxes." Molly sighed and pulled the giant present sack tighter over her shoulder. "On three, we go in, we smile, we steal childhood joy. One... two..." "Do we have to smile? My smile scares kids." "Bill, your everything scares kids. Three."
created_at
"2026-07-25T18:09:30.297Z"
7/25/2026, 6:09:30 PM
model
"meta/muse-spark-1.1-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
7338
moderation_latency
(null)
generation_time
14873
tokens_prompt
75
tokens_completion
1005
native_tokens_prompt
225
native_tokens_completion
1826
native_tokens_completion_images
0
native_tokens_reasoning
905
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
0.00804175
router
(null)
provider_responses
0
endpoint_id
"b2b9f6f9-8880-41c1-bd0c-867650fd5238"
id
"resp_6a64fbda36dbc540ce2541b5"
is_byok
false
latency
326
model_permaslug
"meta/muse-spark-1.1-20260709"
provider_name
"Meta"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.18.0; linux; x64))"
http_referer
(null)
request_id
"req-1785002970-tpaLdhAWS1G3VdDMyFAG"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1785002970-064UgUI2Akune9KVzrlv"
upstream_id
"resp_6a64fbda36dbc540ce2541b5"
total_cost
0.00804175
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Meta"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
690
0.00%
Dialogue to Total Word Ratio
Ratio: 97.69%, Deviation: 27.69%
neededClean
false
wordsTotal
694
wordsDialogue
678
0.0000%