NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37Q09BFYR8J9Z8JQE4XHAHN
Run Details
Model:
Aion 3.5 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
23.7s
Cost:
$0.004909
Provider:
AionLabs
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
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 blueprint 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. That gives us ninety minutes of window before the cake comes out." Bill squinted at the blueprint. "This is a floor plan of a bounce house." "It's a modified bounce house. Janet rented it for the backyard. The presents are going on a folding table beside it. I've done my reconnaissance." "You sat in your car across the street for three hours eating corn dogs." "Reconnaissance," Molly repeated, "comes in many forms." Bill pulled out a chair and sat down heavily, the kind of sigh that suggested he had already lost this argument several times before it began. "Molly. These are children's presents. We're professionals. We once stole a Monet from a hotel lobby during a wine tasting." "And where did that get us? Nobody even noticed for six days. But a child's birthday party? That's intimate. That's personal. That's a heist with heart." "That's a felony involving minors." "Alleged minors. We don't know the ages. Some of those kids look forty." Bill rubbed his temples. "Walk me through it. Just so I can hear how insane it sounds out loud." Molly straightened, delighted. "Phase one: we enter as entertainers. I've got a clown suit." "Absolutely not." "You haven't heard the plan." "The clown suit is disqualifying on its own, Molly. Every heist that's gone wrong in history has involved a clown suit. It's documented. There's a database." "That database is a corkboard in your garage." "A well-organized corkboard." Bill leaned forward. "What's phase two?" "Phase two, we locate the present table. We swap each gift with an identical box filled with rocks." "Where are we getting identical boxes?" "I've been saving them. Since Christmas. It's been a long year, Bill. I've had time." Bill stared at her for a long moment. "You've been wrapping rocks since December." "I wrap when I'm stressed. Don't make it weird." "Okay. Say we get the presents. Say no parent notices two clowns hauling forty boxes to a van shaped like a shark—" "The van is a necessary evil." "—how do we get out of the neighborhood? There's one road in." Molly smiled slowly and produced a walkie-talkie from her jacket. "That's where you come in. You'll be stationed at the entrance in a utility uniform, directing traffic for a fake gas leak." "A gas leak." "A very serious gas leak. Nobody questions a man in a vest, Bill. It's the fundamental weakness of every planned community." Bill opened his mouth, closed it, then opened it again. "What happens when the kids open the rocks?" Molly's smile faltered for the first time all afternoon. "I hadn't... technically gotten there yet." "Eight children, Molly. Eight children opening rocks while a clown sprints through a hedge with a stuffed unicorn under each arm." "When you say it like that, it sounds bad." "It sounds bad in every configuration." There was a pause. Somewhere outside, a lawnmower droned on, indifferent to crime. "New plan," Molly said, snatching the blueprint off the table. "We hit the retirement home raffle instead. Bunch of gift baskets. Nobody runs fast at those places." Bill stood up and grabbed his jacket. "Now that," he said, "is a heist with heart."
created_at
"2026-09-23T18:03:48.481Z"
9/23/2026, 6:03:48 PM
model
"aion-labs/aion-3.5-20260923"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
937
moderation_latency
(null)
generation_time
23656
tokens_prompt
75
tokens_completion
913
native_tokens_prompt
65
native_tokens_completion
810
native_tokens_completion_images
(null)
native_tokens_reasoning
59
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
0.00490875
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790186628-mvkD9RDQePFpiKnLPjQZ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790186628-FzUPEZ6kXdPKGJz5uFk4"
upstream_id
"1860c9ed7aaa4593a355aeea96a67826"
provider_responses
0
endpoint_id
"09b1fc56-3888-42bd-b7a0-926851c7f9d6"
id
"1860c9ed7aaa4593a355aeea96a67826"
is_byok
false
latency
927
model_permaslug
"aion-labs/aion-3.5-20260923"
provider_name
"AionLabs"
status
200
total_cost
0.00490875
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
0.00%
Matches word count
n/a
neededClean
false
words
561
0.00%
Dialogue to Total Word Ratio
Ratio: 73.40%, Deviation: 23.40%
neededClean
false
wordsTotal
564
wordsDialogue
414
0.0000%