NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KRZ639RFCBW062CRJS4XDP
Run Details
Model:
Muse Glimmer 30B (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
23.9s
Cost:
$0.001507
Provider:
Phala
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
4.55%
Dialogue to Total Word Ratio
(details)
2.2770%
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 a crayon drawing of a pink castle across the table and tapped it with a chipped red nail. The kitchen smelled of burnt coffee and cheap pizza. Bill sat opposite her in a stolen bathrobe, wearing a party hat he'd found in a dumpster. A map of the Hendersons' cul-de-sac was taped over a stain. Outside the rain hammered the fire escape. They had three hours until the birthday party started. Molly said the presents would be piled in the living room under a plastic unicorn arch. Bill said the kids would be loud enough to cover a safe crack. "We are not robbing a bank, Bill. We're robbing a seven year old," Molly said. "Same principle, smaller vault. Less alarms, more juice boxes." "There will be parents." "Parents will be busy pretending to like Frozen. Again." Molly pushed the drawing aside and pulled out a list. "Entry through the back fence, through the laundry room window. No glass, just a latch. I checked yesterday when I volunteered for the bake sale." "Volunteered?" "I told Mrs. Henderson my grandma makes perfect cupcakes. She believed me." Bill grinned. "You baked?" "I microwaved frosting and prayed. She cried. We're in." The rain picked up. Bill traced the route with a dry-erase marker. "We go in at twenty past seven, right when they start the piƱata. Everyone's eyes are up. We grab the gift table, split the load between two tote bags." "And the unicorn arch?" "We'll take it too. It's part of the ambiance. Resale value." Molly laughed and then stopped. "Bill, these are kids' presents. They will be cars and dolls and LEGO sets. No resale." "Sentimental value is resale. Remember the Beanie Baby boom." "You are a menace." "I'm a professional menace. What's the exit?" "Front door, through the garage. I disabled the motion sensor with a hairpin last week." "You disabled it with a hairpin?" "It was a heroic act of domestic espionage." Bill stood and stretched his bad knee. "Okay, costumes. We need to blend." "We're going as the ice cream truck and the clown." "We are not." "Clown is you. You already own the shoes." Bill looked down at neon green squeaky shoes. "They match my soul." Molly handed him a paper hat and a plastic badge that said Party Helper. "You will carry the cooler. I will carry the tote. If anyone asks, we're with the balloon animal guy." "Do we have a balloon animal guy?" "We are the balloon animal guy." The clock ticked. Molly zipped the tote bags and checked the window latch one more time. Bill put the party hat straight and practiced his smile in the cracked microwave door. It looked mostly like a threat.
created_at
"2026-09-03T13:57:39.055Z"
9/3/2026, 1:57:39 PM
model
"meta/muse-glimmer-30b-20260810"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
285
moderation_latency
(null)
generation_time
23827
tokens_prompt
75
tokens_completion
1538
native_tokens_prompt
118
native_tokens_completion
1356
native_tokens_completion_images
(null)
native_tokens_reasoning
743
native_tokens_cached
75
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
0.0015075
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788443859-UHKdQP6n9xpXRdPge7Q6"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788443859-8QpCRXLNxUyLq1bHFU2Z"
upstream_id
"req_766756a01d7420758f339488357aa969"
provider_responses
0
endpoint_id
"6244f10e-1e45-488b-a47f-f575e0edd12b"
id
"req_766756a01d7420758f339488357aa969"
is_byok
false
latency
243
model_permaslug
"meta/muse-glimmer-30b-20260810"
provider_name
"Phala"
status
200
total_cost
0.0015075
cache_discount
0.0000195
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
455
4.55%
Dialogue to Total Word Ratio
Ratio: 57.86%, Deviation: 7.86%
neededClean
false
wordsTotal
458
wordsDialogue
265
2.2770%