NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KRWVS4FAAVEP9ZKCA513GJ
Run Details
Model:
Muse Glimmer 30B (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
19.0s
Cost:
$0.001673
Provider:
Fireworks
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
2.08%
Dialogue to Total Word Ratio
(details)
51.0408%
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 was taping a pink plastic tiara to the dashboard with a roll of duct tape and a look of absolute commitment. Bill watched her from the passenger seat, holding a printed invitation that said BILLY'S 7TH BIRTHDAY - NO UNINVITED GUESTS. Bill said, "Molly, this is a children's party. There will be juice boxes and a bouncy castle. Are we sure this is a heist?" Molly said, "It's a heist with extra sugar. Presents, Bill. Dozens of presents. Parents buy the good stuff now. It's a gold mine wrapped in wrapping paper." Bill said, "Last time we hit a kid's party we got caught by a six year old with a foam sword." Molly said, "That six year old was me. And I was on the right side of the law because I was stealing back my own Pokémon cards." Bill said, "Fine. What's the plan?" Molly said, "We go in as party entertainers. I saw it on a flyer. We are The Sparkle Squad. Face paint, balloons, chaos." Bill said, "I am not doing face paint." Molly said, "You are doing glitter beard. It's a disguise. Kids love a glitter beard." Bill said, "Kids love a glitter beard that isn't me." Molly said, "You can be the quiet one. You stand by the snack table and you watch the door. I do the distracting. I will teach the kids the Electric Slide and when they are all on the floor I will sweep the gift table." Bill said, "Sweep the gift table?" Molly said, "Yes. With a tote bag. I have a tote bag that says I LOVE PARENTS. Very ironic." Bill said, "What about the parents?" Molly said, "Parents are on their phones. Parents are getting cake on their shirts. Parents are not a threat. The threat is Timmy with the juice box launcher." Bill said, "Okay. Entry point?" Molly said, "Back gate. The invitation says the address is 14 Maple. The back gate is unlocked. I checked on Instagram. Mrs. Henderson posts her garden parties." Bill said, "You stalked a birthday party host." Molly said, "I am a professional." Bill said, "What if someone opens a present early?" Molly said, "Then we steal it early. Improvisation is key." Bill said, "What if a kid cries?" Molly said, "We give them a balloon animal. I can make a decent poodle." Bill said, "You can't make a decent poodle." Molly said, "Watch me. Also we are in and out in twenty minutes. Presents in the van, glitter off, we are gone before the piñata drops." Bill said, "And if we get caught?" Molly said, "We say we were hired. We have a fake badge. I laminated it." Bill said, "You laminated a fake badge that says Sparkle Squad Certified?" Molly said, "It says Sparkle Squad. Certified Fun. It's perfect." Bill sighed and adjusted the glitter beard. "Fine. But if I get a sticky handprint on my jacket I'm charging you." Molly grinned. "Deal. Let's go steal Christmas in July."
created_at
"2026-09-03T13:56:22.955Z"
9/3/2026, 1:56:22 PM
model
"meta/muse-glimmer-30b-20260810"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
114
moderation_latency
(null)
generation_time
18893
tokens_prompt
75
tokens_completion
1188
native_tokens_prompt
117
native_tokens_completion
1111
native_tokens_completion_images
(null)
native_tokens_reasoning
482
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
0.00167273
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788443782-jxDrD3wGvvSa2h3qXBMz"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788443782-WYrrNZlj6az56d6AfKAD"
upstream_id
"chatcmpl-chatcmpl-bc72196d4fac401b846466cb18150183"
provider_responses
0
endpoint_id
"38b85376-97ef-4ae3-b2f4-b23ca1ee2495"
id
"chatcmpl-chatcmpl-bc72196d4fac401b846466cb18150183"
is_byok
false
latency
21
model_permaslug
"meta/muse-glimmer-30b-20260810"
provider_name
"Fireworks"
status
200
total_cost
0.00167273
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
100.00%
Matches word count
n/a
neededClean
false
words
500
2.08%
Dialogue to Total Word Ratio
Ratio: 78.80%, Deviation: 8.80%
neededClean
false
wordsTotal
500
wordsDialogue
394
51.0408%