NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-07-01KWZ3416TWYRKRBYTFWH97FHR
Run Details
Model:
Aion 3.0
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
5m 39s
Cost:
$0.090363
Provider:
AionLabs
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
88.90%
Dialogue to Total Word Ratio
(details)
44.4487%
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 spread the blueprint across the kitchen table. The overhead light flickered, casting strange shadows across the poorly drawn map. "Okay, so the target is the Henderson house. They've got a six-year-old turning six, and from what I hear, the kid's got a rich uncle who works at Fisher-Price." Bill leaned over the blueprint, his brow furrowed in concentration. "That's not a blueprint. That's a crayon drawing." "The kid drew it. It's surprisingly accurate." Bill squinted at the colorful page. "Is that a dog?" "Security." "A dog is security?" "Buster. Ninety pounds of slobbering security." Bill poured himself a cup of coffee from the pot that had been sitting on the counter since morning. He took a sip and grimaced at the bitter taste. "Can we get past him?" "Of course we can. I've got a plan." "I'm listening." "We go in through the side gate. The dog sleeps in the backyard, so we need to be quiet." "How quiet?" "Ninja quiet." "Last time you said that, you tripped over a garden hose and broke your ankle." "That was different." "How?" "I was wearing flip-flops. This time I'm wearing sneakers." "What about the parents?" "That's the beauty of it. They'll be inside, supervising sugar-crazed children. They won't even notice us." "Until the kids see us taking their presents." Molly shook her head, tapping the crayon drawing with her finger. She traced a path around the side of the house. "That's why we're going in dressed as caterers. Everyone ignores the help." "We don't look like caterers." "We will once I get the outfits." "Where are you getting caterer outfits?" "I've got a guy." Bill set down his coffee mug, the ceramic clanking against the wooden table. He leaned back. "Last time you had a guy, he sold us a getaway car that broke down three blocks from the bank." "That was bad luck." "It was a 1998 Honda Civic with 200,000 miles on it." "Okay, forget the caterer idea. New plan." Bill sat down at the table, his chair creaking under his weight. He crossed his arms. "I'm all ears." "We dress up as mascots." "Mascots?" "The party has entertainment. A clown and a guy in a dinosaur costume. We replace them." "How do we replace them?" "I don't know. Kidnap them?" "Kidnap a clown?" "Knock him out, stuff him in a closet. We walk in, do some juggling, and then slowly walk out with the gifts." Bill stared at her, his eyes wide with disbelief. He let out a long breath. "You want to knock out a clown." "It's a kids' party. He's probably exhausted already." "And the dinosaur?" "Same thing." "So our plan is to assault two children's entertainers, steal their costumes, infiltrate the party, and walk out with all the presents." "Yes." "What if the kids cry?" "They're going to cry anyway when they realize their presents are gone." Bill rubbed his face, exhaling a long breath. He reached into his pocket and pulled out a crumpled pack of cigarettes. "Fine. But if we get caught, I'm blaming you." "Fair enough." "And if the dog bites me, you're paying my medical bills." "Deal. Now help me find a good juggling tutorial on YouTube."
created_at
"2026-07-07T20:08:13.29Z"
7/7/2026, 8:08:13 PM
model
"aion-labs/aion-3.0-20260707"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
610
moderation_latency
(null)
generation_time
338504
tokens_prompt
77
tokens_completion
16444
native_tokens_prompt
67
native_tokens_completion
15027
native_tokens_completion_images
(null)
native_tokens_reasoning
15593
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
0.090363
router
(null)
provider_responses
0
endpoint_id
"8c8f827d-2bef-468a-8f9e-8e27b470de5e"
id
"0e77d8a9f21b4da7b6359edb31a222a3"
is_byok
false
latency
610
model_permaslug
"aion-labs/aion-3.0-20260707"
provider_name
"AionLabs"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.18.0; linux; x64))"
http_referer
(null)
request_id
"req-1783454893-yDheS5zX0sq1QYsg95lb"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783454893-MJ2x6htCMcr4OZ18FjLs"
upstream_id
"0e77d8a9f21b4da7b6359edb31a222a3"
total_cost
0.090363
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
536
88.90%
Dialogue to Total Word Ratio
Ratio: 71.53%, Deviation: 1.53%
neededClean
false
wordsTotal
541
wordsDialogue
387
44.4487%