NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36P58B9VASR5B7D9GZH8H2Z
Run Details
Model:
GPT-6 Luna (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
21.2s
Cost:
$0.000670
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
1.23%
Dialogue to Total Word Ratio
(details)
0.6127%
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 and Bill crouched behind a hedge outside the community hall, where a banner proclaimed HAPPY BIRTHDAY, LEO! in letters large enough to be read from the moon. Through the windows, children shrieked with the tireless joy of people who had never paid rent. A mountain of presents occupied one corner. Bill peered over the hedge. “There must be forty.” “Thirty-eight,” Molly whispered. “I counted the bows.” “You counted the bows?” “I’m a professional.” “You’re a professional bow-counter?” “I’m a professional thief. The bows are incidental.” They had come prepared, in the broad sense of the word. Bill wore a black turtleneck and a paper party hat. Molly carried an empty laundry basket and a clipboard bearing the words DEFINITELY NOT SUSPICIOUS. Bill had suggested a sack, but Molly said sacks made them look like burglars. The clipboard, she explained, made them look like people with authority, which was much worse. “Once we’re inside,” Bill said, “we take every present.” “Quietly.” “Obviously.” “And quickly.” “Of course.” “And no improvising.” Bill looked down at his party hat. “This was your idea.” “It was a disguise.” “It says ‘Birthday Boy’ on it.” “So nobody will suspect you’re here to steal birthday presents.” “Because they’ll suspect I’m the birthday boy?” “They’ll be confused. Confusion is our ally.” From the hall came a burst of music, then a small voice announced that the next game involved a spoon, a potato, and “no running near the cake.” Bill shuddered. “I hate games with rules.” “You once lost at pin the tail on the donkey.” “The donkey moved.” “The poster was taped to a wall.” “It was a very confusing wall.” Molly unfolded her clipboard. The top sheet was blank. Beneath it, she had written a list: ENTER, COLLECT GIFTS, EXIT, BUY SANDWICHES. She crossed out the last item and added, in careful capitals, DO NOT EAT THE EVIDENCE. Bill read it. “What evidence?” “I don’t know yet. That’s why it’s a precaution.” A side door opened, and a woman carrying a tray of cupcakes stepped outside. The thieves ducked. Bill’s party hat caught on a branch and stayed behind, bobbing above the hedge like a tiny festive warning. The woman glanced toward the bushes. “Hello?” Molly stood abruptly, clipboard raised. “We’re here about the presents.” The woman blinked. “You mean the gift table?” “Exactly,” Bill said, popping up beside her. “We’re, uh, quality control.” “Of the presents?” “Children can be very rough with wrapping paper,” Molly added. The woman studied their basket, their matching black clothes, and the word SUSPICIOUS on the clipboard. Then she sighed. “Are you the entertainers?” Bill brightened. “We can be.” “Wonderful. The magician canceled. You’re on in two minutes.” As the woman hurried back inside, Molly stared at Bill. “You said no improvising.” “I didn’t improvise. I accepted a job.” “You can’t do magic.” “I can make presents disappear.” “That is precisely the problem.” Inside, the children began chanting for the magician. Bill squared his shoulders. Molly tucked the clipboard under her arm. “Fine,” she said. “We go in, do one trick, and leave with nothing.” Bill frowned. “Nothing at all?” “Maybe a cupcake.” He nodded gravely. “For morale.” Then they stepped into the party, where thirty-eight brightly wrapped presents waited—and every child in the room was watching.
created_at
"2026-09-23T08:29:48.273Z"
9/23/2026, 8:29:48 AM
model
"openai/gpt-6-luna-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
10514
moderation_latency
(null)
generation_time
21144
tokens_prompt
75
tokens_completion
1064
native_tokens_prompt
66
native_tokens_completion
1327
native_tokens_completion_images
0
native_tokens_reasoning
556
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
0.0006701
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790152188-btwHbnPwoxxNnQEgMGBZ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790152188-nY929CMnwM0Mje36yCyO"
upstream_id
"resp_0bb4f705c865de44016ab38dfc640887d1908007f9eedf1b56"
provider_responses
0
endpoint_id
"05e94e02-b9c2-4bbb-ba55-4082ee9ad687"
id
"resp_0bb4f705c865de44016ab38dfc640887d1908007f9eedf1b56"
is_byok
false
latency
2174
model_permaslug
"openai/gpt-6-luna-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.0006701
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
549
1.23%
Dialogue to Total Word Ratio
Ratio: 39.38%, Deviation: 9.38%
neededClean
false
wordsTotal
551
wordsDialogue
217
0.6127%