NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-07-01KWZ5GZ8X5PRW91YF339ZPMJT
Run Details
Model:
Aion 3.0 Mini
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
42.3s
Cost:
$0.002705
Provider:
AionLabs
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0004%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% 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 and Bill crouched behind a hedge, peering at the birthday party through a white picket fence. A bouncy castle dominated the lawn, its cheerful colors now seeming like a fortress wall. “See the gift table? Stacked ten deep,” Molly said, her voice a low hum. “Too many parents,” Bill said. “We need a distraction.” “I thought of that. The clown. When he does the balloon animal finale, everyone’s watching him. We slip in the back door.” “Back door?” Bill frowned. “I was thinking we take the front. Walk in like we’re late dads.” “That’s your plan? Walk in like dads?” “Late dads,” Bill emphasized. “I got a hat. ‘World’s Best Dad’.” Molly pinched the bridge of her nose. “You don’t have kids, Bill. You’re a convicted thief. No one buys it. We need something better. I got a helium tank. We fill the whole room with balloons.” “For cover?” “No, for a distraction. Everyone looks up as balloons cover the ceiling. We go in low, grab the pile.” Bill nodded slowly. “Then how do we get out? The parents will see us.” Molly pulled out a large, rainbow-striped bag. “I got a custom piñata. It’s full of bugs. We hang it, break it, chaos ensues. We slip out in the confusion.” “That’s… that’s smart. But what about their security? The real dad who thinks he’s a ninja?” “Uncle Greg is in the kitchen, mixing punch. He’s no threat.” Bill squinted. “You got an in with the mom?” “You could say I did a little reconnaissance at the PTA barbecue.” The back door of the house swung open. A man in a party hat staggered out, holding a beer. “Showtime,” Molly whispered. Bill fumbled with the helium tank. The nozzle released a hiss that sounded distressingly like a cat. “Dammit, Bill.” “Sorry. Butterfingers.” They watched a child spot the van and point. Bill said, “What now?” Molly sighed. “We do it the old way. You distract with the piñata, I run for it.” “You gonna beat that thing?” “I’m going to drop it on the birthday boy. Trust me.” Bill held the piñata. “I hate kids’ parties.” Especially when you’re stealing from them?” That too.
created_at
"2026-07-07T20:50:14.443Z"
7/7/2026, 8:50:14 PM
model
"aion-labs/aion-3.0-mini-20260707"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
721
moderation_latency
(null)
generation_time
42189
tokens_prompt
77
tokens_completion
2094
native_tokens_prompt
66
native_tokens_completion
1899
native_tokens_completion_images
(null)
native_tokens_reasoning
1481
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
0.0027048
router
(null)
provider_responses
0
endpoint_id
"0ad7ab68-a351-4a96-b2ef-fa8fe13adae3"
id
"a5281c956e6a4632b6c319be4b67d5bf"
is_byok
false
latency
721
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-1783457414-dPWISaqrqNoSu28YOGR2"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783457414-NyrxSDmcDCotbVF0cpts"
upstream_id
"a5281c956e6a4632b6c319be4b67d5bf"
total_cost
0.0027048
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
365
0.00%
Dialogue to Total Word Ratio
Ratio: 65.30%, Deviation: 15.30%
neededClean
false
wordsTotal
366
wordsDialogue
239
0.0004%