NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36R650JT4JHYC9CX4DZZCCD
Run Details
Model:
GPT-6 Luna (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
1m 45s
Cost:
$0.004841
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.93%
Dialogue to Total Word Ratio
(details)
99.9643%
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
Molly spread a map of the party across the table of the Closed Casket Café. The map was a napkin from the bakery next door, but it showed a cake with alarming accuracy. Bill leaned over it, wearing a paper crown marked BIRTHDAY PRINCE. His mustache curled at the ends like two question marks regretting their punctuation. “We enter at dusk,” Molly said. “The children will be distracted by cake, balloons, and whatever noise a piñata makes when it senses danger. Especially if it has a tiny hat.” “And we leave with every present?” Bill asked. “Every present,” Molly confirmed, “including the suspiciously large one shaped like a pony.” Bill traced the route with a breadstick. Their plan involved no locks, no alarms, and no machinery, because Bill had once mistaken a smoke detector for a tiny moon and apologized to it for three days. Instead, they would walk through the front gate during the party, carrying a gift so convincingly wrapped that even Molly didn’t know what was inside. “It’s probably a toaster,” Bill said. “I wrapped it in the dark. It has a handle, somehow.” Molly inspected the parcel. It had a label reading TO: SOMEONE IMPORTANT, FROM: BILL, and a ribbon tied in the shape of an anxious octopus. She tapped it. The box produced a faint clanking sound, followed by a smaller clank, as if the contents were arguing about the journey. “What is in it?” Molly asked. “A surprise,” Bill said. “For whom?” “The person who opens it.” “That’s usually how gifts work.” “Is the toaster plugged in?” Molly asked. “Not emotionally,” Bill replied. The party began at six in the community hall, a place whose windows displayed paper stars and whose door displayed a handwritten sign: NO GROWN-UPS WITHOUT A CHILD. Molly frowned at the sign. Bill brightened. Their disguise, he explained, was already perfect: they were two grown-ups without a child, which meant the sign had described them exactly. “Could we borrow a child?” Molly whispered. “Where would we get one?” “The orphanage?” “Bill.” “Fine. A cardboard cutout. It won’t ask for a cut.” “What about a puppet, then?” Molly offered. Molly pinched the bridge of her nose. The job was supposed to be simple: arrive, mingle, collect every brightly wrapped box, and depart before anyone noticed that the presents had developed legs. Bill had added the legs to the plan after watching a parade of costumed mascots and deciding camouflage meant being shaped like an animal. “What animal?” Molly asked. “A gift horse,” Bill replied. “We don’t look a gift horse in the mouth.” “We don’t look like one either.” Outside, the first children arrived in a noisy flock, each carrying a balloon or a parent, and all wearing the alert expressions of people who had been promised sugar. Molly watched the door swing open. “On my signal,” Molly said, “we enter.” Bill nodded. “What is the signal?” “I say signal.” “Excellent. I’ll be ready.” He saluted with the breadstick. It snapped.
created_at
"2026-09-23T09:05:14.781Z"
9/23/2026, 9:05:14 AM
model
"openai/gpt-6-luna-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4482
moderation_latency
(null)
generation_time
104943
tokens_prompt
75
tokens_completion
2118
native_tokens_prompt
66
native_tokens_completion
9668
native_tokens_completion_images
0
native_tokens_reasoning
8999
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
0.0048406
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1790154314-S5vmWwWOIQjD5bp664La"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790154314-1EPu7tx8vYtWtYGlesmG"
upstream_id
"resp_0ceab437a0542d1a016ab3964b1a5887d1a0857d08927a5d7b"
provider_responses
0
endpoint_id
"05e94e02-b9c2-4bbb-ba55-4082ee9ad687"
id
"resp_0ceab437a0542d1a016ab3964b1a5887d1a0857d08927a5d7b"
is_byok
false
latency
638
model_permaslug
"openai/gpt-6-luna-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.0048406
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
100.00%
Matches word count
n/a
neededClean
false
words
500
99.93%
Dialogue to Total Word Ratio
Ratio: 29.88%, Deviation: 0.12%
neededClean
false
wordsTotal
502
wordsDialogue
150
99.9643%