NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1K51FNYCV676HQZ1TCWXR1V
Run Details
Model:
Muse Spark 1.3 (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
1m 23s
Cost:
$0.036032
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 stolen birthday invitation across the dashboard of their dented getaway van and tapped the glittery unicorn with a chewed pen. Bill sat beside her wearing a party hat that was far too small and leaking confidence. Rain drummed outside the suburban neighborhood. "We are stealing from a toddler, Bill, so act professional." Bill adjusted the tiny hat and looked deeply offended. "I am professional, I brought balloons and zero guilt." Molly rolled her eyes so hard the van almost tipped and checked her list. "The party starts at three, so we go in as entertainers." Outside children screamed with joy as a rented bouncy castle wheezed in the humid afternoon. "Great, I love kids, they love me, this is perfect cover." Bill watched longingly while Molly sketched the house layout on a napkin stained with ketchup. "No, they do not love you, last time you cried at musical chairs." He opened his mouth to protest, then thought better of it and ate a cold french fry. "That music was emotional, Molly, and Timmy cheated shamelessly." A minivan pulled up carrying more brightly wrapped boxes than either thief had ever seen. "Focus, we grab the presents during cake, then exit through the backyard." Bill whistled low and rubbed his hands together like a cartoon raccoon eyeing trash. "What about the clown, the bouncy castle, and the suspicious mothers?" Molly smacked his hands apart and pointed toward the side gate covered in streamers. "You distract the clown, I will handle mothers with juice boxes." The gate squeaked in the wind, revealing a backyard full of chaos, frosting, and opportunity. "Why do I always get clown duty, they smell like fear and greasepaint?" Bill shuddered at the memory of painted smiles and oversized shoes squeaking on tile. "Because you look like a sad clown already, no costume needed." Molly tossed him a red nose that bounced off his forehead and rolled under the seat. "Rude but accurate, fine, I will juggle badly until you signal." He fished it out, sneezed violently, and immediately looked more ridiculous than before. "Signal is me yelling surprise, then we run with the loot." Across the street the birthday song began, shrill and loud enough to rattle windows. "Perfect plan, toddlers never suspect a thing." They bumped fists over the steering wheel as sprinkles began falling from a broken pinata in the distance. Molly folded the map into her pocket and started the engine, which coughed like an elderly walrus. Bill practiced his juggling with three stale donut holes, dropping every single one onto the floor mats. The plan felt stupid, risky, and strangely perfect, exactly their usual style. Children continued shrieking, parents continued chatting, and the mountain of presents sat unwatched near the fence. Molly grinned widely, Bill adjusted his tiny hat, and both thieves felt the familiar electric thrill of a terrible idea coming together beautifully at last. Tomorrow headlines would either praise their genius or warn parents about very weird entertainers indeed.
created_at
"2026-09-03T08:09:22.885Z"
9/3/2026, 8:09:22 AM
model
"meta/muse-spark-1.3-20260902"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5040
moderation_latency
(null)
generation_time
83193
tokens_prompt
75
tokens_completion
827
native_tokens_prompt
68
native_tokens_completion
8458
native_tokens_completion_images
0
native_tokens_reasoning
7723
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
0.0360315
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788422962-1HblURUw1uH8ztSqkIGB"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788422962-oYBSYgDKl19ExF5MrsB8"
upstream_id
"resp_6a992b3419b115cc5e8f47c2"
provider_responses
0
endpoint_id
"cf4f1b4e-1719-4b65-9111-7dd7635e5a2f"
id
"resp_6a992b3419b115cc5e8f47c2"
is_byok
false
latency
1247
model_permaslug
"meta/muse-spark-1.3-20260902"
provider_name
"Meta"
status
200
total_cost
0.0360315
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%