NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-07-01KWZ53R2Q4935TQA6NQF1KZ09
Run Details
Model:
Aion 3.0
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
18.0s
Cost:
$0.005403
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 30% 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
The fluorescent lights of the twenty-four-hour diner buzzed overhead as Molly spread the crumpled napkin across the formica table. Bill hunched forward, his leather jacket creaking, as he squinted at the crude map she'd drawn in blue ink. "This is the kitchen window," Molly said, tapping a square with her chewed fingernail. "We go in through there at exactly two fifteen, when the clown starts his balloon animal act. Every kid in the place will be focused on a poodle made of latex." "And the parents?" "Wine, Bill. Lots of wine. I overheard the mother at the grocery store. She said, and I quote, 'If I don't get a glass of chardonnay in my hand by two o'clock, I'm going to scream.'" Bill leaned back and rubbed his stubbled chin. "So we're counting on suburban alcoholism to pull this off. Makes me proud to be a criminal." "Stop being dramatic. We're stealing presents, not curing cancer." "What kind of haul are we talking about?" Molly's eyes lit up as she pulled a folded printout from her jacket pocket. "I hacked the Target registry. We're talking Nintendo Switch, three American Girl dolls, a LEGO Death Star, and approximately fourteen pieces of plush garbage. Resale value: four grand, easy." "Four thousand for ruining a child's birthday?" "Children are resilient. They bounce back. Besides, the kid's turning six. She won't remember." "She'll remember if her entire party gets robbed." "Which is why we wait until the cake cutting. Everyone's in the dining room. We hit the garage, where they're staging the gifts. In and out in four minutes." Bill flagged down the waitress and ordered two more coffees. When she left, he leaned in conspiratorially. "What about the hired entertainment? You said there's a clown." "His name is Bonkers." "Of course it is." "Relax. Bonkers is focused on not making crying children. He won't notice us." "And if he does?" "Then we have a forty-year-old man in face paint trying to stop us. I think we can handle it." Bill took a long sip of the coffee that arrived, wincing at the taste. "There's something wrong with this plan. Something I'm not seeing." "You're overthinking it." "I'm not. I'm underthinking it, and even my underthinking is telling me this is stupid." Molly folded up the napkin map and shoved it back in her purse. "We need this. The fence has a buyer for the dolls. We skip this job, we don't make rent." "We could get jobs." "You can't hold a job. You got fired from a gas station." "That register was confusing." "It had pictures of the money on the buttons, Bill." He waved a hand dismissively. "Fine. The kid's party. Two fifteen. Kitchen window." "Garage door, actually. I was lying about the window to throw you off in case you were wearing a wire." "You think I'd wear a wire?" "I don't put anything past a man who can't operate a cash register with pictures on it." Bill drained his coffee and stood up, tossing a crumpled five onto the table. "Two fifteen. I'll bring the van. If this goes sideways, I'm blaming you." "If it goes sideways, Bonkers will probably try to make you a balloon animal. Accept it gracefully."
created_at
"2026-07-07T20:43:01.1Z"
7/7/2026, 8:43:01 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
17723
tokens_prompt
77
tokens_completion
985
native_tokens_prompt
67
native_tokens_completion
867
native_tokens_completion_images
(null)
native_tokens_reasoning
156
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
0.005403
router
(null)
provider_responses
0
endpoint_id
"8c8f827d-2bef-468a-8f9e-8e27b470de5e"
id
"9bbaff1dc6b241509b99c6846420dd60"
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-1783456981-hk0LFL3EChqk5uzHmklT"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783456981-RGcDk1pzMMnRzCS5NqHW"
upstream_id
"9bbaff1dc6b241509b99c6846420dd60"
total_cost
0.005403
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
540
0.00%
Dialogue to Total Word Ratio
Ratio: 75.37%, Deviation: 45.37%
neededClean
false
wordsTotal
544
wordsDialogue
410
0.0000%