NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYD6HXA9BVTKDVY12EET10SV
Run Details
Model:
Muse Spark 1.1 (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
40.4s
Cost:
$0.031043
Provider:
Meta
Evaluation summary
Score
Evaluator
Details Link
9.61%
Matches word count
(details)
90.37%
Dialogue to Total Word Ratio
(details)
49.9886%
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 stolen birthday blueprint across their van hood. The paper showed a bouncy castle and a table labeled PRESENTS with glitter. Bill lowered his backwards binoculars and stared at the suburban house. Juice boxes were stacked like gold bars. "So we're really robbing a seven-year-old's birthday party now?" Bill said. Molly tapped the blueprint with a broken laser pointer that used to be a cat toy. "We're liberating underappreciated assets from an overindulged minor," Molly said. Bill frowned at the present pile drawing. One gift was obviously just socks. "Intel says he's getting a Switch and forty bucks in cards," Molly said. "Where did you get intel from?" Bill asked. "His mom posted the wish list and alarm code on Nextdoor," Molly said. A hot wind blew crepe paper across the street. Somewhere a child screamed with joy. "This is a new low, even for us," Bill said. "This is perfect. Low security, high sugar distraction, guard dog's a goldfish," Molly said. "That says goldfish crackers will be served, not actual goldfish," Bill said. Molly flipped the blueprint upside down and squinted harder. The bouncy castle became a lava pit. "Minor detail. We enter as entertainers. Flawless," Molly said. Bill looked down at his tiny clown suit. It was missing one shoe and smelled faintly of panic. "You want me to be Mr. Giggles again? Kids bite," Bill said. "You kept making balloon tax auditors. Kids want dogs," Molly said. "They requested financial stability and I delivered," Bill said. She traced their route with a chewed pen. Side gate, past juice boxes, straight to presents. "We strike during cake. Vigilance drops when fire appears," Molly whispered. "Then we grab everything and sprint away screaming?" Bill asked. "Walk calmly with the stolen gift wagon. Confidence is everything," Molly explained. Bill's red nose honked when he inhaled nervously. Laughter floated over the fence. "What if the birthday boy cries though?" Bill asked. "Code Blue. You distract with magic," Molly said. "I only know how to hide my thumb. That's just lying convincingly," Bill said. Molly zipped up her discount wizard cape from a retired magician turned accountant. It sparkled aggressively. "Seven-year-olds respect aggressive lying. Trust me," Molly said. Someone started the pinata. The sound of frantic stick swinging echoed. Bill winced sympathetically. "Fine, I'm in. But if caught, we play pin the tail," Bill said. "Deal. Now hand me the cake-scented chloroform," Molly said. "We banned that after the bake sale fiasco," Bill said. Molly sighed and produced a grocery bag full of empty wrapping paper. Recycling was part of the plan. "Operation Birthday Burgle starts in three minutes. Ready, Mr. Giggles?" Molly said. "Born ready. Let's steal childhood innocence and that Lego Death Star," Bill said. They bumped fists. The van door squeaked shut. The bouncy castle loomed like destiny. "One last thing. If we get cake, we eat it before prison," Bill said. "Obviously. No heist works on an empty stomach," Molly said. They marched toward screaming triumphantly. "Remember, act natural. Clowns are supposed to look terrified of children," Bill said. "Perfect. You were method acting all along," Molly said.
created_at
"2026-07-25T17:53:17.647Z"
7/25/2026, 5:53:17 PM
model
"meta/muse-spark-1.1-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
37252
moderation_latency
(null)
generation_time
40325
tokens_prompt
75
tokens_completion
838
native_tokens_prompt
225
native_tokens_completion
7238
native_tokens_completion_images
0
native_tokens_reasoning
6490
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
0.03104275
router
(null)
provider_responses
0
endpoint_id
"b2b9f6f9-8880-41c1-bd0c-867650fd5238"
id
"resp_6a64f80d5272780ccf7d413b"
is_byok
false
latency
289
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-1785001997-dimKhPVhRdlKnNUxnmXH"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1785001997-JW0rlGdr5vfy4pkz1tor"
upstream_id
"resp_6a64f80d5272780ccf7d413b"
total_cost
0.03104275
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
9.61%
Matches word count
n/a
neededClean
false
words
522
90.37%
Dialogue to Total Word Ratio
Ratio: 48.58%, Deviation: 1.42%
neededClean
false
wordsTotal
527
wordsDialogue
256
49.9886%