NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36P48KA540GNXW2CHNCSZQ2
Run Details
Model:
GPT-6 Luna (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
16.3s
Cost:
$0.000616
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.08%
Matches word count
(details)
99.70%
Dialogue to Total Word Ratio
(details)
49.8916%
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
“Tonight,” Molly said, “we become legends.” Bill peered over the garden fence. “The kind of legends who steal presents from children?” “The kind who steal presents from children and then feel complicated about it.” “That’s not a legend. That’s a lengthy apology.” Behind the fence, a banner announced HAPPY BIRTHDAY, LEO! Through the windows came the thump of music and the bright, shrieking laughter of a dozen small people. Molly adjusted her black hat. “We go in, collect every present, and leave before anyone notices.” “How do we collect every present?” “Quickly.” “And leave?” “Also quickly.” “Excellent. I’ll add that to my notes under ‘details.’” “You brought notes?” “I brought a shopping list. I crossed off milk and wrote ‘moral flexibility.’” Molly sighed. “Bill, we’ve discussed this. We need the money.” “We need money for rent. The children need presents for being children. I’m not sure whose need wins, but I suspect the children have a stronger case.” “Don’t get sentimental. It ruins your face.” “My face is already ruined. I ate an expired yogurt to save money.” They watched a small boy march past the window wearing a cardboard crown. He was carrying a plush dinosaur nearly as large as himself. Bill lowered his voice. “That one looks happy.” “He’s probably the birthday boy.” “Then he’s got a lot to lose. A crown, a dinosaur, possibly a future in local government.” “We can’t steal his future. We’re stealing the gifts.” “Those are often related.” Molly folded her arms. “Do you have a better plan?” “Yes. We go in, say we’re the surprise entertainers, and perform a show. Then we leave without taking anything.” “That’s not a heist.” “It’s a heist where we steal snacks.” “Bill, no one hires two strangers in black coats to entertain children.” “Correct. We’ll have to be convincing.” “You can’t juggle.” “I can juggle three objects.” “Last week you dropped two eggs and a shoe.” “The shoe was unexpected.” A woman opened the back door and called, “Are you the party helpers?” Molly and Bill froze. Bill whispered, “This is our moment.” “Our moment to run?” “Our moment to say yes.” Molly stared at him. “You want to perform?” “I want to avoid being the villains in a story told by a child with excellent memory.” The woman waved them over. “The magician’s late!” Molly glanced at Bill. “You’re the magician.” “I’m the what?” “Your notes say moral flexibility.” “That doesn’t mean I own a cape.” “It means you’re flexible. Go.” Inside, the children gathered around as Bill produced a coin from behind one boy’s ear. The boy gasped. Molly clapped as if the coin had been a tiger. “Now,” Bill announced, “for my greatest trick: I will make this birthday cake disappear.” “No!” shouted the children. “Fair,” he said. “I’ll make the vegetables disappear.” Molly leaned toward him. “We’re not stealing the presents?” “Not tonight.” “Because of the children?” “Because the cake is enormous, and I’ve just discovered my true calling.” “Magician?” “Unpaid cake consultant.” Molly looked at the pile of gifts, then at the children’s shining faces. “Fine. But I’m taking two slices.” “Thief.” “Consultant,” she said, and reached for a plate.
created_at
"2026-09-23T08:29:15.76Z"
9/23/2026, 8:29:15 AM
model
"openai/gpt-6-luna-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
6726
moderation_latency
(null)
generation_time
16249
tokens_prompt
75
tokens_completion
1203
native_tokens_prompt
66
native_tokens_completion
1218
native_tokens_completion_images
0
native_tokens_reasoning
476
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
"completed"
service_tier
"default"
usage
0.0006156
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790152155-bkkepz8a0wMSNXPEeaxm"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790152155-QZFBJ9BN5aMlDLEUGk9g"
upstream_id
"resp_0b383c453ed5ed6a016ab38ddbda9087d181de6923e6e8e488"
provider_responses
0
endpoint_id
"05e94e02-b9c2-4bbb-ba55-4082ee9ad687"
id
"resp_0b383c453ed5ed6a016ab38ddbda9087d181de6923e6e8e488"
is_byok
false
latency
1322
model_permaslug
"openai/gpt-6-luna-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.0006156
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
0.08%
Matches word count
n/a
neededClean
false
words
529
99.70%
Dialogue to Total Word Ratio
Ratio: 69.75%, Deviation: 0.25%
neededClean
false
wordsTotal
529
wordsDialogue
369
49.8916%