NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36R9WBPK3NBBEYWY41BSYG0
Run Details
Model:
GPT-6 Luna (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
21.0s
Cost:
$0.000745
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
32.22%
Dialogue to Total Word Ratio
(details)
16.1101%
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 hand-drawn map across the dashboard. It showed the community hall, the cake table, and a large red X over a box labeled PRESENTS. Bill leaned closer, fogging the windshield with his breath. Outside, balloons bobbed above the hall’s windows, bright as warning flags. “Tonight, we take every present,” Molly said. “All of them. A clean sweep.” “Do we know how many?” “Enough to make a very small sleigh.” Bill studied the map. “And what’s the security?” “Children. Parents. One magician.” “Ah,” he said. “The magician is the problem.” “He’s six years old.” “Children are unpredictable. They have sticky hands and no respect for a good disguise.” Molly tapped the map. The party was for a boy named Oliver, who was turning seven. His parents had hired a clown, a magician, and a woman who made balloon animals. Bill considered this an excessive amount of surveillance for a birthday. Their plan, Molly explained, was to enter as entertainers, blend into the festivities, and leave with the presents while everyone was distracted. Bill would perform a magic trick. Molly would handle the presents. The trick was meant to buy her time. “What’s my trick?” Bill asked. “You make a coin disappear.” “Where does it go?” “Into your pocket.” “That’s not magic. That’s theft with a flourish.” “Good. You’re already in character.” Bill frowned at the map. “What if the magician challenges me?” “Then you challenge him back.” “To what?” “Something he can’t do.” “Like pay taxes?” Molly sighed. “Just make the coin disappear.” They had acquired costumes from a bargain shop: Bill wore a satin cape that said THE AMAZING BILLY, and Molly wore a floppy hat decorated with felt stars. The hats were supposed to make them look whimsical. On Bill, the hat made him look like a mushroom accused of fraud. He practiced the coin trick in the rearview mirror. The coin dropped down his sleeve, bounced off his elbow, and landed in his tea. “Perfect,” Molly said. “No one will suspect you.” “I suspect me.” A burst of laughter floated from the hall. Through the window, they saw children gathered around a table, where a mountain of wrapped gifts glittered beneath paper streamers. One enormous package wore a bow nearly as wide as Bill’s shoulders. “That one’s mine,” he whispered. “It’s for Oliver.” “I meant emotionally.” Molly folded the map and tucked it into her coat. “Remember: in, presents, out. We don’t touch the cake.” “Why not?” “Because last time you touched the cake, you got stuck in it.” “It was a very soft cake.” They climbed from the car. Bill adjusted his cape and tried to look magical. Molly straightened his felt-star hat. “Ready?” she asked. “Ready,” Bill said. “Though if the children ask, I’m a birthday wizard.” “And if the magician asks?” “I’m a birthday wizard with a coin.” Together, they marched toward the hall, where a small boy spotted Bill’s cape, pointed, and shouted, “Look! The mushroom magician’s here!” Bill gave Molly a solemn nod. “Our cover is excellent.” Then the doors swung open, and a dozen children rushed out to greet them, each carrying a present. Molly looked at the packages. Bill looked at the children. “New plan?” he murmured. “Absolutely,” she said. “We steal the magician’s hat.” It was the first sensible idea either of them had had all evening.
created_at
"2026-09-23T09:07:16.996Z"
9/23/2026, 9:07:16 AM
model
"openai/gpt-6-luna-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
7895
moderation_latency
(null)
generation_time
20924
tokens_prompt
75
tokens_completion
1068
native_tokens_prompt
66
native_tokens_completion
1476
native_tokens_completion_images
0
native_tokens_reasoning
722
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
0.0007446
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1790154436-KXY58pXYhyYEu4sDKpMt"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790154436-Et07Qvf086PMZE4Ax5Op"
upstream_id
"resp_0e300b61f8d2c6e5016ab396c519f087d198bda3bd8cb800e5"
provider_responses
0
endpoint_id
"05e94e02-b9c2-4bbb-ba55-4082ee9ad687"
id
"resp_0e300b61f8d2c6e5016ab396c519f087d198bda3bd8cb800e5"
is_byok
false
latency
803
model_permaslug
"openai/gpt-6-luna-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.0007446
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
559
32.22%
Dialogue to Total Word Ratio
Ratio: 34.76%, Deviation: 4.76%
neededClean
false
wordsTotal
561
wordsDialogue
195
16.1101%