NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1K61DRDTYD982TDBK7J2THZ
Run Details
Model:
Muse Spark 1.3 (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
1m 38s
Cost:
$0.022482
Provider:
Meta
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
42.14%
Dialogue to Total Word Ratio
(details)
21.0703%
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
The community center storeroom smelled like old pizza and bleach while Molly spread the birthday party floor plan across a stack of juice boxes. Bill crouched beside her wearing a stolen magician cape that still had rabbit fur on it. Outside they could hear children screaming the birthday song off key for little Tyler who was turning seven and apparently rich and spoiled. Molly whispered, "Okay, Tyler's mom posted the present table on Instagram. It's by the exit, next to the cake. Twenty plus gifts, all wrapped huge. We go in during musical chairs." Bill whispered, "Musical chairs? That's genius. Nobody watches presents during musical chairs. They just watch kids shove each other for a tiny chair." Molly nodded and tapped the wobbly circle she had drawn for the table. "Exactly. You dress as the clown. I already rented you the rainbow wig and the squeaky nose." Bill groaned and scratched at the cape. "Why am I always the clown? Last time I was the clown at the retirement home and I still have glitter in places glitter should never be." Molly grinned, "Because you have natural clown energy, sweetie. Big feet, sad eyes, questionable life choices." Bill sighed, "Fair. So I waddle in with balloon animals and scream surprise? Then what?" Molly tapped the map, "Then you do the world's worst magic show. Pull a juice box out of your sleeve. Cry a little. The parents will be too embarrassed to look away." Bill said, "I can do bad magic. I once made my rent money disappear. So while I suffer, you grab the loot?" Molly smoothed her messy bun and held up a stained coffee mug. "Better. I go as a tired mom. Cardigan, messy bun, mug that says Mommy Needs Wine. No one questions a tired mom at a kids party." Bill laughed, "You stole my aunt's whole vibe. So you just walk out with a sack of Legos?" Molly whispered, "I bring the big reusable Costco bag. Say Tyler forgot his allergy meds and I have to grab his gift bag. Moms help moms. It's perfect camouflage." Bill frowned, "What if they count presents? Rich moms count presents like casinos count chips." Molly said, "They won't. Right after musical chairs is pinata time. Every adult will be filming vertically and yelling swing harder, Tyler." Bill rubbed his jaw at the memory. "Pinata chaos is real. I lost a tooth to a pinata in ninety eight. Still worth it for the Skittles." Molly said, "Focus, Bill. While they film, I slide the presents out the back door by the bathrooms. You meet me with the van." Bill frowned, "The van still says Flowers For Funerals on the side. Is that suspicious for a kids party?" Molly said, "It's perfect. No one steals a funeral flower van. It's invisible with sadness." Bill whispered and pointed toward the gym, "What about the bouncy castle guard? That teenager looks serious. He has a whistle and unresolved anger." Molly laughed, "Brandon? I give him twenty bucks and tell him his girlfriend is here. He will abandon that castle so fast." Bill grinned, "Cold. I like it. But what do we do with twenty presents for a seven year old? If we get a Nintendo Switch I call dibs." Molly folded the map and shoved it into the Costco bag. "If we get a Switch we sell it and pay rent, you clown. No more gas station hot dogs for dinner. Operation Birthday Grab is on." Bill stood and squeaked his red nose. "Fine. Clown suit, fake mom, funeral van, crying magic. What could go wrong?" Molly smiled, "Everything, probably. But if we pull this off we eat real pizza tonight and not dumpster bagels."
created_at
"2026-09-03T08:26:49.493Z"
9/3/2026, 8:26:49 AM
model
"meta/muse-spark-1.3-20260902"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4412
moderation_latency
(null)
generation_time
97333
tokens_prompt
75
tokens_completion
907
native_tokens_prompt
68
native_tokens_completion
5270
native_tokens_completion_images
0
native_tokens_reasoning
4424
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
0.0224825
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788424009-6vPpIZacWcW4N6Ggbl5h"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788424009-sfHTrzPY9kvcnAYvODIL"
upstream_id
"resp_6a992f4a2009b2e1a4364090"
provider_responses
0
endpoint_id
"cf4f1b4e-1719-4b65-9111-7dd7635e5a2f"
id
"resp_6a992f4a2009b2e1a4364090"
is_byok
false
latency
1500
model_permaslug
"meta/muse-spark-1.3-20260902"
provider_name
"Meta"
status
200
total_cost
0.0224825
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
0.00%
Matches word count
n/a
neededClean
false
words
623
42.14%
Dialogue to Total Word Ratio
Ratio: 74.16%, Deviation: 4.16%
neededClean
false
wordsTotal
623
wordsDialogue
462
21.0703%