NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-07-01KWZ54EH9JMFKQB7FH864G6K3
Run Details
Model:
Aion 3.0
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
20.0s
Cost:
$0.005943
Provider:
AionLabs
Evaluation summary
Score
Evaluator
Details Link
43.38%
Matches word count
(details)
0.04%
Dialogue to Total Word Ratio
(details)
21.7103%
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 leaned forward, almost knocking over her coffee. "Here's the deal. Little Timmy Henderson's turning eight on Saturday, and his parents just dropped three grand on a bouncy castle." Bill scratched his stubbled chin. "Three grand? For something kids just jump in?" "Right? And that's not even counting the presents. We're talking Lego sets, gaming consoles, action figures still in their packaging. A treasure trove." Bill squinted. "So we just walk in and take everything?" "No, genius, we need a plan. I was thinking we could disguise ourselves as entertainment." "Like what? Clowns?" Molly shuddered. "God no. Have you ever tried doing anything in those shoes? No, I was thinking face painters." "Face painters don't do anything. They just sit there painting butterflies on chubby cheeks." "Exactly. Perfect cover. You set up your little station by the gift table, and I'll mingle with the parents, casing the layout, maybe distracting them with stories about my fictional kids." Bill nodded slowly. "Which one of us is the face painter?" "You, obviously." "Why obviously?" "Because you have that artistic sensitivity." "You mean I look like the type who'd paint butterflies on chubby cheeks?" "Among other things." Bill grunted. "Fine, but I'm not wearing a smock." "No one asked you to wear a smock. But you'll need to look the part. Maybe a colorful vest." "I draw the line at a colorful vest." "Fine, no colorful vest. You'll just be the most boring face painter in history." "Works for me. So while I'm painting butterflies—" "Stars, actually. Stars are faster." "While I'm painting stars, what are you doing?" "I told you, mingling. And when everyone's distracted by the cake cutting, you grab the gifts and slip them out through the side gate." "What if a kid catches me?" "You're a grown man, Bill. Use your words." "What if a parent catches me?" "Same principle." "What if they call the cops?" "Then we run. Ditch the face painting kit and run." Bill leaned back, arms crossed. "You know, when I got into crime, I imagined bank vaults and laser grids. Not stealing toys from eight-year-olds." "Crime is crime. And kids are resilient. They'll get over it. Remember, we're not hurting anyone. We're just... redistributing wealth." "To ourselves." "To ourselves, yes. Think of it as an early lesson in disappointment for them. Life's going to teach them that eventually anyway." Bill chewed his thumbnail. "I don't know, Molly. This feels different." "It's not different. It's exactly the same. A job is a job. And this one pays in Lego sets." Molly spread out the rough blueprint she'd drawn of the Henderson backyard on the table between them, a simple map that nonetheless captured all the key elements: the bouncy castle location, the gift table, the cake station, and crucially, the side gate that led to the alley where their van would be waiting. "The party starts at two. We arrive at three, when things are in full swing. Less attention on newcomers." Bill pointed at the blueprint. "What's this?" "The piñata station." "Are we stealing the piñata too?" Molly sighed. "Focus, Bill."
created_at
"2026-07-07T20:43:24.088Z"
7/7/2026, 8:43:24 PM
model
"aion-labs/aion-3.0-20260707"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
487
moderation_latency
(null)
generation_time
19829
tokens_prompt
77
tokens_completion
1130
native_tokens_prompt
67
native_tokens_completion
957
native_tokens_completion_images
(null)
native_tokens_reasoning
280
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
0.005943
router
(null)
provider_responses
0
endpoint_id
"8c8f827d-2bef-468a-8f9e-8e27b470de5e"
id
"7acb6ba0ec074e19b24772b25bb7d7aa"
is_byok
false
latency
487
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-1783457004-0FSEi9WdwGfGwCY88HDt"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783457004-g7knLOMzAeot0HaWMNoi"
upstream_id
"7acb6ba0ec074e19b24772b25bb7d7aa"
total_cost
0.005943
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
43.38%
Matches word count
n/a
neededClean
false
words
517
0.04%
Dialogue to Total Word Ratio
Ratio: 82.47%, Deviation: 12.47%
neededClean
false
wordsTotal
519
wordsDialogue
428
21.7103%