NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KRRW7D12BXGHWQ1665V0DY
Run Details
Model:
Muse Glimmer 30B (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
48.0s
Cost:
$0.004479
Provider:
Fireworks
Evaluation summary
Score
Evaluator
Details Link
97.63%
Matches word count
(details)
5.96%
Dialogue to Total Word Ratio
(details)
51.7940%
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
Molly spread the invitation on the kitchen table next to a half-eaten toast. The glitter glue was still wet. Bill peered over her shoulder with a spoon in his mouth. "Okay, listen. It's seven-year-old Timmy's birthday party. Saturday, three to five, at the community center. Theme is Space Explorers." "Do they have presents?" Bill asked. "Bill, it's a children's party. They have presents. There's going to be a table covered in wrapping paper, bows, and parental guilt." He lowered the spoon. "That's the best kind of target. No security, no alarms, just moms with juice boxes." Molly tapped the address with a red fingernail. "We go in at four-thirty. Parents are distracted by cake. Kids are doing the piƱata. We grab the presents, we go." "And what if Timmy cries?" Bill said. "Timmy will cry when he realizes his presents are in our van. That's emotional damage we can live with." Bill laughed. "You got a costume?" "I'm going as a volunteer party helper. Tutu, hairnet, name tag that says Molly. It's perfect cover. You be the bored teenager with a clipboard." "I am a bored teenager with a clipboard," Bill muttered. Molly rolled her eyes. "You need to be less bored and more helpful. You know how to act normal around kids, right?" "I once babysat my nephew for twenty minutes. He tried to feed the cat a cupcake. I think I'm qualified." "Great. Your job is to distract the adults. Ask about the venue rental, the insurance, whether they need more napkins. Bore them into compliance." Bill nodded slowly. "And you grab the loot." "We grab the loot. All of it. The Lego sets, the art kits, the embarrassing stuffed animals with googly eyes. We leave nothing." "What about the parents who spent forty dollars on a single remote control car?" "We donate it to a good cause. Like us." Molly pulled out a small duffel bag. "We have the exit route. Back door by the kitchen. The loading dock is unlocked during events for catering." "And if someone stops us?" "We say we're collecting for the hospital. Kids in space need space toys. It's poetic." Bill stood up and stretched. "This is either the dumbest heist we've ever done or the smartest." "It's both," Molly said, grinning. "That's why I love it." "Fine. But if I get glitter in my hair, I'm charging you." "You'll thank me when we're driving away with a trunk full of plastic dinosaurs." He picked up the invitation. "Timmy's mom wrote 'no uninvited guests.'" Molly snatched it back. "She didn't say no highly qualified party helpers." The toast was cold now, but the plan was warm. Bill grinned and tapped the bag. "What time do we meet?" Molly checked her watch. "Three-thirty in the parking lot. Bring the clipboard, and for the love of God, don't eat the cake before we steal it." "No promises," Bill said, and they both laughed. The heist was on.
created_at
"2026-09-03T13:54:12.34Z"
9/3/2026, 1:54:12 PM
model
"meta/muse-glimmer-30b-20260810"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
124
moderation_latency
(null)
generation_time
47910
tokens_prompt
75
tokens_completion
3015
native_tokens_prompt
117
native_tokens_completion
2982
native_tokens_completion_images
(null)
native_tokens_reasoning
2291
native_tokens_cached
112
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
0.00447923
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788443652-zFXaJJrPRr1k4JKeTCwA"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788443652-VjmsjJu0dryqIIXoKYXq"
upstream_id
"chatcmpl-chatcmpl-7e88d4ebb4544ca7bf4558cbc464a17a"
provider_responses
0
endpoint_id
"38b85376-97ef-4ae3-b2f4-b23ca1ee2495"
id
"chatcmpl-chatcmpl-7e88d4ebb4544ca7bf4558cbc464a17a"
is_byok
false
latency
32
model_permaslug
"meta/muse-glimmer-30b-20260810"
provider_name
"Fireworks"
status
200
total_cost
0.00447923
cache_discount
0.00003472
upstream_inference_cost
0
provider_name
"Fireworks"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
97.63%
Matches word count
n/a
neededClean
false
words
493
5.96%
Dialogue to Total Word Ratio
Ratio: 77.51%, Deviation: 7.51%
neededClean
false
wordsTotal
498
wordsDialogue
386
51.7940%