NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1K5NRMH8CDSQR48G64VZHHN
Run Details
Model:
Muse Spark 1.2 (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
25.6s
Cost:
$0.024093
Provider:
Meta
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.95%
Dialogue to Total Word Ratio
(details)
99.9746%
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 crumpled community center floor plan across the getaway car's hood while Bill tried to untangle a balloon arch from the antenna. "Okay, so Timmy's seventh birthday starts at two and every single present gets piled on the magician's table by the juice boxes." Bill popped the last balloon with a satisfying bang and stared at the drawing as if it might bite him. "We're really doing this? Stealing from seven year olds feels like a new low, even for us." "Don't be dramatic. It's not stealing from kids, it's liberating overpriced plastic from ungrateful kids who already have too many toys." She tapped a red X marked near the back entrance, next to a scribbled note that read CAKE. "That's our way in. The kitchen door sticks. The staff props it open for the pizza delivery guy." Bill leaned closer, squinting at her handwriting, which looked dangerously like a treasure map drawn by a toddler. "And the exit strategy? We just waddle out carrying a mountain of Paw Patrol wrapping paper?" "We wear the uniforms. I ordered two extra large clown costumes. Nobody questions a clown carrying presents." Bill held up the costume bag. A giant red nose fell out and bounced across the pavement. "Molly, I have clown trauma. My cousin Barry chased me with a balloon poodle for three hours." She ignored him completely and pulled out a list titled Timmy's Loot, with items like LEGO pirate ship underlined three times. "Focus. The parents will be watching the magician saw a lady in half. That's when we grab the table." The wind lifted the blueprint and Bill chased it halfway down the parking lot, wrestling it away from a startled pigeon. "What if Timmy cries? I can't handle a kid crying. My resolve crumbles faster than that princess castle cake." "Then you do the funny voice. You know, the one that makes toddlers laugh and old ladies tip extra." Bill practiced his squeaky clown voice, which sounded more like a deflating air mattress than entertainment. "Like this? Honk honk, little Timmy, your presents are now property of two very desperate adults?" Molly doubled over laughing, nearly dropping the loot list into a puddle of melted ice cream. "Perfect. Absolutely terrifying. The kids will love it. Let's move before the face painter finishes and blocks our door." They squeezed into the ridiculous costumes behind a dumpster, adjusting wigs and oversized shoes while a mom wheeled a wagon of soda past without glancing twice. "Do I look distinguished or just desperately sweaty in polka dots?" Molly honked his nose for him and grinned, already hearing the faint sound of children singing happy birthday inside. "Definitely distinguished. Now grab the decoy cake and try not to trip." Bill lifted the cardboard cake, wobbling on enormous shoes as Molly flung open the sticky kitchen door to reveal a mountain of glittering presents waiting beyond. Neither thief noticed the sleeping clown already inside. "Is that snoring really part of the plan?"
created_at
"2026-09-03T08:20:27.415Z"
9/3/2026, 8:20:27 AM
model
"meta/muse-spark-1.2-20260805"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
24603
moderation_latency
(null)
generation_time
25595
tokens_prompt
75
tokens_completion
763
native_tokens_prompt
68
native_tokens_completion
5658
native_tokens_completion_images
0
native_tokens_reasoning
5005
native_tokens_cached
35
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
0.024093
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788423627-GHVdi7uuGdyoVVgkHZgR"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788423627-IBtYG2TxROgFWbCvahLl"
upstream_id
"resp_6a992dcf225a088c76144435"
provider_responses
0
endpoint_id
"ffc8e76e-808a-44d2-bd21-c6ac16e721cb"
id
"resp_6a992dcf225a088c76144435"
is_byok
false
latency
3784
model_permaslug
"meta/muse-spark-1.2-20260805"
provider_name
"Meta"
status
200
total_cost
0.024093
cache_discount
0.0000385
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
501
99.95%
Dialogue to Total Word Ratio
Ratio: 50.10%, Deviation: 0.10%
neededClean
false
wordsTotal
501
wordsDialogue
251
99.9746%