NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36P3CFWFH4EVDRR02T2D4Q9
Run Details
Model:
GPT-6 Luna (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
16.2s
Cost:
$0.000587
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.48%
Dialogue to Total Word Ratio
(details)
0.2413%
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
“Children’s parties,” Molly said, unfolding a napkin on the café table, “are the perfect place to steal presents.” Bill studied the napkin. “You’ve drawn a cake.” “It’s a planning diagram.” “It has candles.” “Those indicate danger.” “They indicate birthday.” Molly tapped the paper. “We arrive during the excitement, collect every wrapped gift, and disappear before anyone notices.” “Children notice everything. Last week one spotted me eating a biscuit from across a park.” “You were holding the biscuit.” “He had excellent instincts.” Molly lowered her voice. “We need a disguise.” “Clown?” “Too memorable.” “Magician?” “Too suspicious.” “Uncle?” Molly paused. “That’s disturbingly plausible.” Bill leaned back. “I could wear a cardigan and say, ‘How you’ve grown.’” “No one at a children’s party has grown. That’s why they’re children.” The café door opened, and a small girl marched in carrying a cardboard crown. Her parents followed, balancing a lopsided cake. Bill watched them. “Is that the party?” “Not unless the celebration is for surviving public transportation.” The girl stopped at their table. “Are you the people stealing the presents?” Molly froze. Bill looked at his hands. “Why would you think that?” Molly asked. “Because you’re talking about stealing presents.” Bill nodded. “She has excellent instincts.” The girl pointed at Molly’s napkin. “That’s not a map. It’s a cake with a weather forecast.” Molly glanced down. A smear of coffee had spread across the candles. “It’s raining.” “Are you bad people?” the girl asked. “Professionally,” Bill said. Molly kicked him under the table. “Only in a broad, theatrical sense,” she said. “We’re reconsidering.” The girl climbed onto the empty chair. “My birthday’s next week. I don’t want anyone stealing my presents.” “Of course not,” said Molly. “Then why are you planning it?” “Because,” Bill said, “we’re very bad at choosing hobbies.” The girl considered this. “You could help with the party instead.” Bill frowned. “What would that involve?” “Carrying things. Making balloon animals. Keeping my cousin away from the frosting.” Molly folded the napkin. “That sounds like a complicated operation.” “It’s harder than stealing presents,” the girl said. “The balloons keep escaping.” Bill looked at Molly. “She’s got us there.” “We’d need a team,” Molly said. “A distraction.” “I can sing,” said the girl. “Too powerful,” Bill replied. “We’d attract the entire neighborhood.” Her father set the cake down. “You two know each other?” “Not yet,” Molly said. “We’re applying for work.” The girl handed Bill her cardboard crown. “You can be the balloon person.” He placed it on his head. It fit badly and gave him the solemn appearance of a king who had lost a small war. Molly sighed. “So we’re not stealing the presents?” “No,” Bill said. “We’re guarding them.” “From whom?” He glanced at the girl. She grinned. “My cousin.” Molly stood. “Then we’ll need a plan.” Bill picked up the cake diagram. “Finally, a job with candles.” “They’re danger indicators,” Molly reminded him. “Good,” he said. “I’m excellent with danger.” The girl took his hand and led him toward the door. “First, you have to learn the balloon animals.” Bill looked back at Molly. “If I make a giraffe, does it count as a getaway vehicle?” “Only if it’s very fast,” she said. The cardboard crown slid over his eyes. “Then I may need a smaller horse.”
created_at
"2026-09-23T08:28:46.978Z"
9/23/2026, 8:28:46 AM
model
"openai/gpt-6-luna-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
7225
moderation_latency
(null)
generation_time
16203
tokens_prompt
75
tokens_completion
1016
native_tokens_prompt
66
native_tokens_completion
1160
native_tokens_completion_images
0
native_tokens_reasoning
364
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
0.0005866
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790152126-1mP15mViNzDxkpd5kdJg"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790152126-2CGERnCVeTOxtxeiayqN"
upstream_id
"resp_0bfa2ce579dc1eb1016ab38dbf11c887d1adb362b2255e837f"
provider_responses
0
endpoint_id
"05e94e02-b9c2-4bbb-ba55-4082ee9ad687"
id
"resp_0bfa2ce579dc1eb1016ab38dbf11c887d1adb362b2255e837f"
is_byok
false
latency
1716
model_permaslug
"openai/gpt-6-luna-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.0005866
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
0.00%
Matches word count
n/a
neededClean
false
words
548
0.48%
Dialogue to Total Word Ratio
Ratio: 59.67%, Deviation: 10.33%
neededClean
false
wordsTotal
548
wordsDialogue
327
0.2413%