NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYD6BMJB9D5ZGT4Q4EAVVHPN
Run Details
Model:
Muse Spark 1.1 (Reasoning, Minimal)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
17.1s
Cost:
$0.006429
Provider:
Meta
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.10%
Dialogue to Total Word Ratio
(details)
0.0520%
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 blueprint across the laundromat folding table. It wasn't a bank blueprint. It was a flyer for Timmy's Super Dinosaur Birthday Bash, laminated and sticky with what she hoped was fruit punch. Bill squinted at it through his reading glasses, which made him look less like a hardened criminal and more like a disappointed substitute teacher. "This is a new low," he said. Molly tapped a glitter-covered T-rex with a chipped nail. The whole operation was marked in crayon. Bounce house by the oak. Cake table by the patio. Present mountain, circled three times in red, located dangerously close to a clown. "We have principles, Bill. We don't do guns, we don't do banks with armed guards, and we don't do jobs that require pantsuits," Molly said. "Stealing from a seven-year-old was also on the list," Bill muttered. "That was you. You added that after the library kid beat you at chess," Molly said. Bill adjusted his cap. He had spent twenty years learning how to crack safes, hotwire cars, and now he was studying a backyard with a pinata that could fight back. "The Johnsons always go overboard," Molly continued. She smoothed the flyer. "Last year the kid got a genuine gold-plated gaming console, a pony, and savings bonds. Savings bonds, Bill. Just sitting in envelopes. Like little paper bricks of cash." Bill leaned in. The present table was indeed marked PRESENT MOUNTAIN and next to it someone had drawn what appeared to be a fire-breathing lizard guarding it. Or maybe that was the aunt. "Security?" he asked. "One dad filming everything on an iPad, one mom named Linda who has already had three mimosas, and a magician who only knows how to pull scarves out of his sleeve," Molly said. Bill nodded slowly. This was his process. "Entry point is the side gate," he said. "It squeaks. Everyone will hear it." "So we WD-40 it," Molly said. "It's in the plan. We enter as entertainment," Bill said. Molly frowned. "I am not wearing the clown suit again." "You wore it to the DMV heist and it worked perfectly," Bill said. "It itched and the nose honked every time I lied," Molly said. Bill pulled two small bags from his coat pocket. One said CANDY. One said DECOY CANDY. "We go as parents who brought extra juice boxes. No one questions a tired adult with juice boxes. We mingle, we compliment the cake, we locate the mountain," he explained. Molly grinned. The grin was why Bill had followed her into retirement savings fraud, a botched aquarium robbery, and now this. "What about the kids?" she asked. "We distract them. You do your disappearing quarter trick. Kids love that," Bill said. "My disappearing quarter trick ends with me keeping the quarter," Molly said. "Exactly. It's method acting for this job," Bill replied. The plan was, in its own terrible way, perfect. They would arrive at three fifteen during the dinosaur stomp dance contest, when all adult supervision was legally abandoned. Bill would engage Linda in conversation about property values. Molly would deploy the emergency Capri-Sun spill near the present table, causing a sticky evacuation. Then, in the chaos, they would wheel the entire plastic wagon of gifts straight out the side gate and into their van, which they had already decorated with a sign that read TIMMY'S DINOSAUR EXPRESS. Molly rolled up the flyer. "One question," she said. "Shoot," Bill said. "If we get caught, do we have to stay for the party games?" Bill considered this, the full weight of his criminal career pressing on his shoulders. "Worse," he said. "We have to help clean up."
created_at
"2026-07-25T17:49:52.081Z"
7/25/2026, 5:49:52 PM
model
"meta/muse-spark-1.1-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
10578
moderation_latency
(null)
generation_time
16952
tokens_prompt
75
tokens_completion
913
native_tokens_prompt
225
native_tokens_completion
1463
native_tokens_completion_images
0
native_tokens_reasoning
644
native_tokens_cached
64
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
0.0064286
router
(null)
provider_responses
0
endpoint_id
"b2b9f6f9-8880-41c1-bd0c-867650fd5238"
id
"resp_6a64f7462cc032fccf534484"
is_byok
false
latency
6028
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-1785001792-XDAAnCAtYkuxBqEdo7eB"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1785001792-z52JPovfjUv0nWeGEdOo"
upstream_id
"resp_6a64f7462cc032fccf534484"
total_cost
0.0064286
cache_discount
0.0000704
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
609
0.10%
Dialogue to Total Word Ratio
Ratio: 41.72%, Deviation: 11.72%
neededClean
false
wordsTotal
616
wordsDialogue
257
0.0520%