NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1K4PM2KWCDG8T32A4ETSHEV
Run Details
Model:
Muse Spark 1.2 (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
41.6s
Cost:
$0.029954
Provider:
Meta
Evaluation summary
Score
Evaluator
Details Link
99.98%
Matches word count
(details)
96.59%
Dialogue to Total Word Ratio
(details)
98.2857%
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 minivan smelled like stale french fries and destiny. Outside the Jelly Bean Palace, a banner that read TIMMY TURNS SIX flapped crookedly in the wind. Molly lowered her binoculars. Bill tugged at his clown suit. "Okay, walk me through the plan again without using the word heist for a six-year-old's birthday," Bill said. "We go in, we secure the assets, we get out, it's a classic heist," Molly said. "The assets are action figures and crayons wrapped in dinosaur paper," Bill said. "High value assets, limited edition, plus there's probably a bike," Molly said. "Molly, there's a clown already inside, he'll know I'm fake," Bill said. "He's a teenage clown making minimum wage, he wants to be anywhere but here, you're method acting," Molly said. "I can't make balloon animals, last time I tried I popped three and a child cried," Bill said. "Kids love tragedy, just hand them the popped remains and call it modern art," Molly said. "And how do you get the presents? There's a mom guarding that table like a dragon," Bill said. "I'll go through the back, past the ball pit, while you do your big distracting clown finale," Molly said. "My finale? I don't have a finale, my act is sweating and squeaking," Bill said. "Perfect, kids love physical comedy, fall over near the cake and they'll swarm you," Molly said. "What if I get stuck in the bouncy castle? I'm not as agile as I was," Bill said. "Then I will rescue you after I secure the mountain of presents, priorities, Bill," Molly said. "This is the worst heist ever, we're stealing juice boxes and goodie bags," Bill said. "And cake, don't forget cake, grab me a slice with extra frosting," Molly said. Molly flipped open a notebook covered in glitter stickers. Inside was a diagram drawn in crayon with arrows pointing to the present table, the cake, and an escape route labeled WINDOW. "You drew the plan in crayon?" Bill said. "You said blend in, I borrowed supplies from the craft table," Molly said. Bill stared at the entrance where a mother herded screaming children inside. The clown's shoes squeaked loudly. He looked down at his own giant red shoes. "What about the parents? They will definitely notice two adults stealing gifts," Bill said. "They'll be too busy filming the piƱata, that's our window, literally," Molly said. "Fine, but if we get caught I'm saying you kidnapped me and forced me into clownery," Bill said. "Deal, now remember, when I yell surprise, you yell distraction and fall over," Molly said. "And if Timmy cries?" Bill said. "We give him back one present and say it's magic, he'll buy it," Molly said. "Last chance to back out, we could just go get tacos and pretend this never happened," Bill said. "Are you kidding? The frosting alone is worth the felony, plus I already wrote Timmy's name off the gift tags," Molly said. "Let's steal that cake before the kids eat it all," Molly said.
created_at
"2026-09-03T08:03:27.101Z"
9/3/2026, 8:03:27 AM
model
"meta/muse-spark-1.2-20260805"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2712
moderation_latency
(null)
generation_time
41360
tokens_prompt
75
tokens_completion
759
native_tokens_prompt
68
native_tokens_completion
7028
native_tokens_completion_images
0
native_tokens_reasoning
6339
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
0.029954
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788422607-C67tM6q3mV8jx9nRJprU"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788422607-apGayFmQ7oP2XITaLxOu"
upstream_id
"resp_6a9929cf8bf8d1fabdc44f1a"
provider_responses
0
endpoint_id
"ffc8e76e-808a-44d2-bd21-c6ac16e721cb"
id
"resp_6a9929cf8bf8d1fabdc44f1a"
is_byok
false
latency
414
model_permaslug
"meta/muse-spark-1.2-20260805"
provider_name
"Meta"
status
200
total_cost
0.029954
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
99.98%
Matches word count
n/a
neededClean
false
words
502
96.59%
Dialogue to Total Word Ratio
Ratio: 70.83%, Deviation: 0.83%
neededClean
false
wordsTotal
504
wordsDialogue
357
98.2857%