NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KRA90Y25NYBDX6PDAHFAEX
Run Details
Model:
Muse Glimmer 30B (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
17.1s
Cost:
$0.001249
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 spread the flyer across the kitchen table like it was a blueprint for the Louvre. The paper showed a pink plastic castle, a banner that read LILY TURNS SEVEN, and a very helpful map of the cul-de-sac with a big red X on 14B. She tapped the X with a pen cap. Bill peered over her shoulder, still in his pajama pants and a reflective vest he claimed was for night jobs. He sniffed the air and asked if she had finally started baking again. She had not. The smell was stale coffee and the faint chemical tang of the air freshener she sprayed to cover it. She pointed to the time on the flyer, three to six, cake at four thirty. She pointed to the presents table under the patio awning. Bill whistled. He had been in houses, banks, museums, but never a backyard with a bouncy castle. He said the bounce was a problem. The noise was a problem. The parents were a problem. Mostly the kids. Molly said the kids were the best part. She had watched a birthday party from the hedges last month. Once the music started and the pizza arrived, the adults stopped watching the door. They watched the sugar crash. Bill asked how they were getting in. Molly said the back fence was lattice, easy to see through, easier to climb. She had already cased it on Tuesday while walking the dog she did not own. The presents would be on folding tables near the grill, unwrapped, unguarded, and tagged with names in glitter pen. He said glitter pen was a trap. It gets everywhere. She said that was why she liked it. No one would notice a few extra hands in the chaos. They would need a bag, two bags, and a story for the exit. She had a story. Bill leaned in. He wanted the story. Molly told him she was volunteering as a balloon animal lady. She had bought a neon pink wig and a certificate she printed at home that said Certified Fun Engineer. Bill said that was a lie. She said it was a cover. The plan was simple and terrible. Arrive at three forty five, just before the chaos peaked. Molly would work the line, making dogs and swords while Bill mingled with dads holding beers. At four twenty they would both slip to the gift table. Bill would lift from the left, Molly from the right. They would stash the haul in a laundry basket covered with a towel. Bill asked what about Lily. Molly said Lily would be busy blowing out candles. He asked what about the parents. She said the parents would be busy taking photos and arguing about who forgot the ice. He said it felt wrong. She said stealing from corporations felt wrong until the money was in their hands. Stealing from children felt wrong until they remembered children outgrow toys in a week and the parents just wanted the mess gone. Bill sighed and reached for the flyer. He folded it neatly and put it in his pocket. He said fine, but if a kid cries he was blaming her. She said she would bring extra balloons for cover. He said he would bring the basket. Outside the window the streetlights flickered on. Somewhere a sprinkler hissed. Molly wrote BALLOONS on a spare tote bag in sharpie. Bill wrote STOLEN PRESENTS on the bottom of the basket in case they got confused.
created_at
"2026-09-03T13:46:13.926Z"
9/3/2026, 1:46:13 PM
model
"meta/muse-glimmer-30b-20260810"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
519
moderation_latency
(null)
generation_time
17003
tokens_prompt
75
tokens_completion
1276
native_tokens_prompt
118
native_tokens_completion
1131
native_tokens_completion_images
(null)
native_tokens_reasoning
418
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
0.00124908
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788443173-8dH5Qaue9YG6Yv59w86a"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788443173-ymj9ZmBFrTcZMSganhTt"
upstream_id
"req_eaa3c8fda3cf7badc7c9f3f6a44ad623"
provider_responses
0
endpoint_id
"6244f10e-1e45-488b-a47f-f575e0edd12b"
id
"req_eaa3c8fda3cf7badc7c9f3f6a44ad623"
is_byok
false
latency
420
model_permaslug
"meta/muse-glimmer-30b-20260810"
provider_name
"Phala"
status
200
total_cost
0.00124908
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
582
0.00%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 30.00%
neededClean
false
wordsTotal
584
wordsDialogue
0
0.0000%