NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-24-01M39X6RM0DNXBP5WHPQAJJ0PP
Run Details
Model:
Aion 3.5 Mini (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
4m 47s
Cost:
$0.020561
Provider:
AionLabs
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
71.32%
Dialogue to Total Word Ratio
(details)
85.6593%
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 van sat in the far corner of the Sunnydale Community Center parking lot, wedged between a minivan and a Prius covered in family stickers. Through the windshield, Molly watched a banner reading "Happy 6th Birthday, Timmy!" flap in the breeze like a taunt. The smell of buttercream frosting drifted through the vents, thick enough to taste. Inside that building sat forty pounds of shrink-wrapped loot—plastic castles, toy dragons, and enough gift cards to fund their retirement in Boca Raton. Retirement had been the plan. Retirement, or at least a very long vacation. Bill adjusted the hot dog costume crumpled in his lap. The bun was made of felt. The wiener was foam. He looked like a man who had lost a bet with a deli. He'd argued for the ninja disguise. Molly had vetoed it, citing "historical inaccuracies" and "the fact that ninjas don't sell hot dogs." It was a compromise that satisfied absolutely no one. "I still don't see why we couldn't rob a jewelry store," he said. "This feels wrong." "Wrong? We stole a Fabergé egg from a nun." Molly spread the blueprints across the dashboard, weighting the corners with coffee cups. "The kid's parents own three car dealerships. This isn't a heist, it's a redistribution of wealth." "A redistribution with a bouncy castle." "The bouncy castle is a distraction. The real problem is the pony." Bill went pale. "The pony?" "Sir Galahad. Retired police horse." She tapped a red circle on the diagram. "He can smell fear. And corn dogs." "That's a very specific smell combination." "That's why you're wearing the costume." She traced the route with her finger. Back entrance, through the kitchen, past the ice cream table. Ninety seconds, in and out, before anyone noticed the presents were gone. The plan was elegant. Simple. Foolproof, which was exactly what worried her, because Bill was involved. Last month he'd set off a museum's alarm system by sneezing. Twice. He slumped in his seat, the foam wiener drooping against the dashboard. The felt bun flopped over his forehead like a sad hat. "If I get trampled by a police pony, I want it noted that I died doing something stupid." "Noted. Now remember, you're Frank the Frankfurter. I'm Aunt Carol." "Nobody will believe you're an aunt." "I have a cardigan and a purse full of butterscotch candies. I am extremely believable." Molly slid out of the van and smoothed her wig. The afternoon sun glinted off the community center's windows. Somewhere inside, a clown was already twisting balloons into shapes that looked less like dogs and more like existential crises. The clown was sweating. She checked her watch. Timmy would blow out his candles in twenty minutes. After that, the presents would be stacked by the door for easy loading—easy for her, anyway. Bill pulled the hot dog hood over his head, breathing through the felt. "If that pony eats me, tell my ex-wife she was right about everything." "That's the spirit, Frank."
created_at
"2026-09-24T14:30:41.036Z"
9/24/2026, 2:30:41 PM
model
"aion-labs/aion-3.5-mini-20260923"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1614
moderation_latency
(null)
generation_time
286887
tokens_prompt
75
tokens_completion
12975
native_tokens_prompt
65
native_tokens_completion
14678
native_tokens_completion_images
(null)
native_tokens_reasoning
12227
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
0.0205609
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1790260241-P8ImSWnRPc36QtgDUMOC"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790260241-XGicjfuTQxclMtE1B0N0"
upstream_id
"ea1f2645b06a4ce1a0356f38fbd07c52"
provider_responses
0
endpoint_id
"60ac3bdc-90ec-4769-9193-2c00b4d79562"
id
"ea1f2645b06a4ce1a0356f38fbd07c52"
is_byok
false
latency
1614
model_permaslug
"aion-labs/aion-3.5-mini-20260923"
provider_name
"AionLabs"
status
200
total_cost
0.0205609
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
100.00%
Matches word count
n/a
neededClean
false
words
499
71.32%
Dialogue to Total Word Ratio
Ratio: 32.60%, Deviation: 2.60%
neededClean
false
wordsTotal
500
wordsDialogue
163
85.6593%