NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37PJ6V7AT84DP0HJ668WFD7
Run Details
Model:
Aion 3.5 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
26.4s
Cost:
$0.008425
Provider:
AionLabs
Evaluation summary
Score
Evaluator
Details Link
98.71%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.3562%
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 spread the map across the kitchen table, weighting the corners with a bag of gummy bears and a half-eaten sandwich. "Okay," she said, tapping the paper. "The party starts at two. Cake at three. That's our window." Bill squinted at the map. "This is a map of a bounce house." "It's the interior layout. I got it from the rental company." "You got a floor plan of a bounce house." "Recon, Bill. It's called recon." Bill ate a few gummy bears and set the bag back down. "I still don't like the clown thing." "You're not doing the clown thing. I'm doing the clown thing." "You can't do the clown thing. You're terrified of clowns." "That's why it's perfect. Nobody suspects a clown who's scared of herself." "That's not how fear works." Molly jabbed a finger at the map. "The presents get stacked on a table next to the piñata. Piñata's at three-fifteen. Every kid in that yard will be losing their minds trying to beat a donkey full of Snickers. That's when we move." "What about the parents?" "Magician's at three-thirty. Parents always watch the magician. They paid forty bucks for him. They want their money's worth." Bill considered this. "That's actually not terrible." "I know." "What about the birthday kid? Kids cry, Molly. Kids cry loud." "That's why you're on distraction duty." "I thought I wasn't doing the clown thing." "You're not. You're doing balloon animals." Bill stared at her. "I don't know how to make balloon animals." "You'll figure it out. It's a tube and air. How hard can it be?" "You said that about the zip line." "The zip line worked." "The zip line dropped you into a koi pond." "Recon pond." Bill stood, paced to the window, and looked out at the perfectly ordinary suburban street where, in forty-eight hours, they would commit the lowest-stakes crime of their careers. Somewhere out there was a folding table soon to be piled with wrapped boxes containing a game console, a telescope, and what Molly insisted was "a robot arm thing." "What if a kid sees us?" he asked. "Then you make them a balloon animal." "I don't know how to make balloon animals." "Then you make them a snake and call it a worm. Kids love worms. Kids are the easiest marks in the world." Bill sighed the sigh of a man who had once stolen a Monet and was now planning around nap schedules. "Fine. But if this goes wrong, I'm telling everyone the clown made me do it." "The clown is going to make herself do it," Molly said, folding the map with the reverence of a sacred document. "Now get your shoes. We're going to the party store. You need practice tubes." "Do I at least get a costume?" "You get a fanny pack and a name tag that says Gary." "Who's Gary?" "Gary is a man who makes balloon animals and asks no questions," Molly said, grabbing her keys. "Be Gary, Bill. Be Gary with your whole heart."
created_at
"2026-09-23T17:56:07.163Z"
9/23/2026, 5:56:07 PM
model
"aion-labs/aion-3.5-20260923"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
746
moderation_latency
(null)
generation_time
26343
tokens_prompt
75
tokens_completion
1515
native_tokens_prompt
65
native_tokens_completion
1396
native_tokens_completion_images
(null)
native_tokens_reasoning
785
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
0.00842475
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790186167-5itbxN45JmMWhVXT4iHT"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790186167-hHw3PdO9QLbxrenn8GO7"
upstream_id
"764a8b991ebb4196afe81b3ef4da9be5"
provider_responses
0
endpoint_id
"09b1fc56-3888-42bd-b7a0-926851c7f9d6"
id
"764a8b991ebb4196afe81b3ef4da9be5"
is_byok
false
latency
746
model_permaslug
"aion-labs/aion-3.5-20260923"
provider_name
"AionLabs"
status
200
total_cost
0.00842475
cache_discount
0.00014625
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
98.71%
Matches word count
n/a
neededClean
false
words
506
0.00%
Dialogue to Total Word Ratio
Ratio: 70.25%, Deviation: 20.25%
neededClean
false
wordsTotal
511
wordsDialogue
359
49.3562%