NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KRAJSF2WDDWGC1HVSQ8S2F
Run Details
Model:
Muse Glimmer 30B (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
34.0s
Cost:
$0.002428
Provider:
Phala
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
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 taped a flyer to the fridge with a crayon drawing of a castle and the words Timmy's 7th Birthday Party Saturday 2pm Backyard. Underneath she wrote Presents inside. Bill stared at it over a chipped mug of instant coffee that tasted like regret. That's a lot of plastic swords and glitter slime, he said. That's a lot of plastic swords and glitter slime we could sell for exactly zero dollars, Molly said. We could sell it to the parents for their emotional damage, Bill said. The doorbell rang. It was Mrs. Henderson with a casserole. Molly opened it in a cape made from a dish towel. We are not taking food, Bill whispered. We are taking presents, Molly whispered back. The planning happened on a kitchen table covered in takeout menus and a stolen blueprint of the Hendersons' backyard from the county assessor's website. The fence was three feet high. The pool was empty. The party was Saturday at two. The sun would be high and cruel. Molly had already cased it. Timmy wanted a dinosaur. His mother had bought seventeen dinosaurs. She had also bought a bounce house, which was a problem because the bounce house would be full of six year olds with no impulse control and sticky hands. There was also a table of presents wrapped in paper that looked like it had been chosen by a committee of well meaning aunts. We go in at three, Bill said. Cake time. Parents are distracted. Kids are sugar crashed. Kids are sugar crashed and crying, Molly corrected. That's perfect cover. She laid out the gear. A pink tutu for camouflage. A prop clipboard that said Party Safety Patrol. A wagon with a hand painted sign that said Candy Delivery. A pair of cheap sunglasses that made Bill look like he was perpetually under investigation. Bill held up a pair of foam devil horns. Why do I need horns, he asked. Because you look like an accountant without them, Molly said. The plan was simple. Walk in as late guests. Bring a fake gift. Blend. Wait for the present table to be unattended during the piñata. The piñata was a T-Rex. Molly had watched YouTube videos. Piñata duty always took twelve minutes and involved a lot of yelling and at least one parent with a broom. Then we grab, Bill said. Then we grab everything, Molly said. Don't be modest. She had a spreadsheet. Presents were estimated at forty seven items. Mostly cheap, some decent. A LEGO set. Two tablets. A remote control car that actually worked. One glittery unicorn costume that weighed more than Bill. We could be millionaires, Bill said. We could be arrested, Molly said. Same thing in this town, Bill said. The tutu was itchy. The clipboard was heavy. The wagon squeaked. Molly tested it in the hallway and it sounded like a mouse with a death wish. We need WD-40, she said. We need a new life, Bill said. Molly smiled. She liked the plan. It was stupid and loud and exactly the kind of thing that would get them featured on the local news under Cautionary Tales. She liked that the target was seven years old and completely defenseless against capitalism. She checked the flyer again. Backyard. Two pm. Presents inside. Fine, Bill said. Let's rob a seven year old. Fine, Molly said. Let's rob a seven year old's future.
created_at
"2026-09-03T13:46:23.925Z"
9/3/2026, 1:46:23 PM
model
"meta/muse-glimmer-30b-20260810"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
260
moderation_latency
(null)
generation_time
33934
tokens_prompt
75
tokens_completion
2440
native_tokens_prompt
118
native_tokens_completion
2203
native_tokens_completion_images
(null)
native_tokens_reasoning
1479
native_tokens_cached
117
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
"stop"
service_tier
(null)
usage
0.00242828
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788443183-Wewl24Utm9fWIdC8vGXZ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788443183-kdcAtD5jNccrCdYj5VDq"
upstream_id
"req_98f4391276b0527a59be5205e915a677"
provider_responses
0
endpoint_id
"6244f10e-1e45-488b-a47f-f575e0edd12b"
id
"req_98f4391276b0527a59be5205e915a677"
is_byok
false
latency
221
model_permaslug
"meta/muse-glimmer-30b-20260810"
provider_name
"Phala"
status
200
total_cost
0.00242828
cache_discount
0.00003042
upstream_inference_cost
0
provider_name
"Phala"
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
568
0.00%
Dialogue to Total Word Ratio
Ratio: 1.23%, Deviation: 28.77%
neededClean
false
wordsTotal
569
wordsDialogue
7
0.0000%