NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-24-01M39WX2FBAQAXH8F9CHP9WN3X
Run Details
Model:
Aion 3.5 Mini (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
3m 1s
Cost:
$0.013004
Provider:
AionLabs
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
99.08%
Dialogue to Total Word Ratio
(details)
49.5379%
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
The hideout was a masterclass in squalor. Located above a laundromat that had been “closed for renovations” since the Reagan administration, it smelled of mildew, gun oil, and old pizza. A single bare bulb hung from the ceiling, casting long shadows over their meager equipment: a crowbar, a sack of water balloons, and a walkie-talkie set that only picked up polka music. Molly stared at the layout of the suburban backyard, her brow furrowed in intense concentration. The blueprint, if one could call it that, was scrawled on the back of a grease-stained pizza menu they’d stolen from a rival gang of vegetarian anarchists. It detailed the enemy territory with clinical precision. A plastic slide molded to look like a dinosaur. A sandpit currently occupied by a family of aggressive ants. A folding table groaning under the weight of a sheet cake decorated with enough blue frosting to cause a national health crisis. She tapped a pen against her teeth, analyzing the tactical advantages of the garden gnome placement. The gnomes were positioned near the rose bushes, providing excellent cover for a tactical advance, though their ceramic smiles were unnerving. Bill was currently eating the cheese puffs that had fallen onto the card table earlier that evening. He looked up, crumbs dusting his patchy beard, his eyes glazed with a mix of boredom and artificial cheddar. “Alright, listen up,” Molly said, slapping the paper flat with her palm. “The target is Timmy’s fifth birthday. We hit them at 2:15 PM, right when the sugar crash starts to set in.” Bill was attempting to balance a fork on his nose. He let it clatter to the table and looked up, crumbs dusting his patchy beard. “A kid’s party? Feels a bit low-level for the likes of us.” “It’s not low-level, Bill. It’s high-yield. Parents go crazy for these things. I heard there’s a Power Wheels Jeep in the haul.” Bill stopped chewing. He looked genuinely impressed, which for Bill was a rare feat of cognitive function. “A Jeep? For a five-year-old? That’s a big investment for a kid who can’t reach the pedals.” “Exactly. And it’s going to be sitting right there on the grass, unguarded, while everyone sings ‘Happy Birthday’ to a kid who can barely read. It’s the perfect score.” Bill nodded slowly, wiping his hands on his jeans. “What about the defenses? I heard they hired entertainment.” Molly grimaced, a look of pure disgust crossing her features. “A clown.” The color drained from Bill’s face. He looked like he was going to be sick. “A clown? Mols, you know how I feel about clowns. They are ancient entities of pure chaos. They defy logic.” “It’s a guy named Dave,” Molly said flatly. “He owes me money. I can handle him.” Bill didn’t look convinced. He picked up his ski mask and turned it over in his hands, examining the holes for signs of wear. “And the extraction? How do we get the goods out?” “The dog door,” Molly replied, pointing to a small rectangle on the map. “Is there a dog?” “A golden retriever. Named Biscuit. He’s friendly. I have a steak.” Bill picked up the steak Molly had brought, weighing it in his hand like a professional gambler checking his cards. The raw meat was cold and red, a stark contrast to the pastel decorations on the map. He didn’t ask where she got it; in their line of work, it was better not to know. Some mysteries were best left unsolved, especially when they involved butcher shops and missing invoices. Bill took a deep breath, squaring his shoulders as if preparing for battle. He pulled the ski mask over his head, getting it momentarily stuck on his nose before popping it free with a grunt. “Okay. Okay. Let’s do this. For the Jeep.” Molly smiled, a cold, sharp expression that promised violence to anyone who stood between her and a battery-operated ride-on toy. “For the Jeep, Bill. For the Jeep.”
created_at
"2026-09-24T14:25:23.45Z"
9/24/2026, 2:25:23 PM
model
"aion-labs/aion-3.5-mini-20260923"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1615
moderation_latency
(null)
generation_time
180839
tokens_prompt
75
tokens_completion
9225
native_tokens_prompt
65
native_tokens_completion
9280
native_tokens_completion_images
(null)
native_tokens_reasoning
8213
native_tokens_cached
65
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
0.0130037
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1790259923-DixqzqZm2mzy1MgHL4qS"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790259923-NHVksGXrvoELJ6CaYtyY"
upstream_id
"7b5a3b181eb5499bb12a38cdf3df4aa2"
provider_responses
0
endpoint_id
"60ac3bdc-90ec-4769-9193-2c00b4d79562"
id
"7b5a3b181eb5499bb12a38cdf3df4aa2"
is_byok
false
latency
1615
model_permaslug
"aion-labs/aion-3.5-mini-20260923"
provider_name
"AionLabs"
status
200
total_cost
0.0130037
cache_discount
0.0000338
upstream_inference_cost
0
provider_name
"AionLabs"
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
664
99.08%
Dialogue to Total Word Ratio
Ratio: 29.57%, Deviation: 0.43%
neededClean
false
wordsTotal
673
wordsDialogue
199
49.5379%