NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-07-01KWZ482QFTE70YSRRK8KHJ8J1
Run Details
Model:
Aion 3.0 Mini
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
1m 15s
Cost:
$0.003287
Provider:
AionLabs
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
7.77%
Dialogue to Total Word Ratio
(details)
3.8864%
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 peered through the binoculars from the passenger seat of the rusted-out panel van. Across the manicured lawn, a bouncy castle deflated with a sad, flatulent wheeze. A banner reading “Happy 5th Birthday, Timmy!” flapped in the breeze. “That,” Bill whispered, pointing a greasy finger, “is our target.” “It’s a birthday party, Bill. For a five-year-old,” Molly said. “Presents, Molly. Pure, unadulterated, unopened potential. Think about it. No security. No vaults. The owners are distracted, mainlining caffeine and trying to stop a kid from eating a worm.” Molly lowered the binoculars. “And the presents are in full view. On a folding table. Covered in cartoon unicorns.” “Exactly. The audacity of it is our in. I’ve got the plan. Phase one: the juice box seduction. I approach the perimeter, disguised as a friendly clown. I offer free balloon animals.” “You can’t make balloon animals,” Molly said. “I have a book! ‘Fun with Latex and Lung Power.’ And a pump. A manual pump, less conspicuous. I make the animals, hand them out. The parents relax. The children are enchanted. Distraction achieved.” “Then what?” “Phase two: the retrieval. You, Molly, will be in the ventilation system.” Bill pointed at the house’s roof. “Ventilation? For a single-story ranch house with central air. That vent is twelve inches wide. And it’s fake. It’s nailed to the soffit.” “It’s a euphemism for ‘devious method of ingress.’ Fine. You’ll be under the cake table. I’ve sewn a camouflage blanket. It has little flowers on it. You are now a bush.” “A floral bush. Under a folding table. While thirty children have their sugar rush.” “It’s touch-and-go for a minute. The toddler with the lollipop almost blows our cover, but you freeze. He moves on. I subtly nudge the table with my clown shoe. The presents slide into your waiting sack.” “What sack?” Molly asked. “The burlap sack I’ll hand you while pretending to tie my shoe. We rendezvous by the piñata at the signal.” “Which is?” “The sound of a child weeping. I’ll tell the birthday boy his dad ate his cake.” Molly stared at him. “That’s monstrous.” “It’s the chaos we need! Momentum! We grab the bag, vault the fence, and we’re gone. In and out, three minutes. Forty gifted, gift-wrapped presents.” “For a five-year-old. The presents are going to be socks. And coloring books. And a plastic toy that screams when you squeeze it.” “Untraceable. And we can sell the screaming toy online. ‘Gently used, mildly cursed.’ Listen, Molly, this is a low-risk, high-annoyance heist. Perfect for our skill set.” A small child with a face smeared in chocolate appeared at the fence. He was holding a wrapped box. “Are you a clown?” the child asked. Bill froze. “Blast. Counter-surveillance.” Molly sighed. “Yes,” she said. “He’s a clown. An idiot clown. He’s leaving now.” “Don’t you want a present?” the kid asked, holding out the box. Bill looked at Molly. His eyes were wide. “It’s a trap,” he whispered. “It’s a kindergartner, Bill.” “A very cunning kindergartner. He’s using the box as bait.” The child tugged at the van’s door handle. “My mom says you need to move your dump truck.” Molly started the engine. Bill fumbled with his seatbelt. “Abort, abort!” he hissed. “The mark is too clever!” As they peeled away, Molly turned to Bill. “Next heist. You pick. And it better have a real ventilation shaft, or I’m finding a new partner.”
created_at
"2026-07-07T20:27:54.497Z"
7/7/2026, 8:27:54 PM
model
"aion-labs/aion-3.0-mini-20260707"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
689
moderation_latency
(null)
generation_time
74644
tokens_prompt
77
tokens_completion
2489
native_tokens_prompt
66
native_tokens_completion
2315
native_tokens_completion_images
(null)
native_tokens_reasoning
1519
native_tokens_cached
0
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
0.0032872
router
(null)
provider_responses
0
endpoint_id
"0ad7ab68-a351-4a96-b2ef-fa8fe13adae3"
id
"7d62e23bb8bc438aa21a89f22de02d67"
is_byok
false
latency
689
model_permaslug
"aion-labs/aion-3.0-mini-20260707"
provider_name
"AionLabs"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.18.0; linux; x64))"
http_referer
(null)
request_id
"req-1783456074-fBgzxivpQiTngTOWaSrr"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783456074-pPq5Dj88hIVhjKJcNgYo"
upstream_id
"7d62e23bb8bc438aa21a89f22de02d67"
total_cost
0.0032872
cache_discount
(null)
upstream_inference_cost
0
provider_name
"AionLabs"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
570
7.77%
Dialogue to Total Word Ratio
Ratio: 77.15%, Deviation: 7.15%
neededClean
false
wordsTotal
582
wordsDialogue
449
3.8864%