NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-09-01KX403RHR6HJY4RXN0SM8BAH8
Run Details
Model:
GPT-5.6 Luna (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
7.8s
Cost:
$0.006060
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
27.17%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
13.5829%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% 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
Rain ticked against the bakery windows while Molly spread a hand-drawn floor plan across a table sticky with frosting. The plan showed the community hall, the buffet, three balloons, and a terrifyingly large question mark beside the restroom. “We enter during the magician’s act,” Molly whispered. “Everyone will be watching him pull scarves from his sleeve.” “Excellent,” said Bill. “Nothing distracts a crowd like a man revealing his shocking lack of laundry.” Molly tapped the paper. “We cross behind the curtain, collect the presents, and leave through the kitchen.” Bill studied the map upside down, then rotated it ninety degrees. “What if the magician is actually competent?” The hall next door throbbed with cheerful music. A banner reading HAPPY BIRTHDAY, LUCY! sagged above the entrance, one corner clinging to a light fixture. Through the windows, children in paper crowns chased one another around a table loaded with cake, juice, and gift-wrapped treasure. Molly adjusted her black wool hat. It had a pom-pom on top, which defeated the entire purpose of looking mysterious. “We’re not stealing from children,” Bill said, peering through the glass. “We’re liberating presents from an environment where they’re being dangerously underappreciated.” “They’re sitting on a table.” “Exactly. Neglected.” A shriek erupted inside, followed by applause. The magician had apparently produced a rabbit. The rabbit looked furious and wore a tiny vest. Molly pulled the door open. “Remember, we’re cousins.” “Which cousins?” “The kind nobody invites twice.” They slipped inside beneath a cloud of balloons. Molly immediately collided with a woman carrying twelve paper plates. Bill caught nine, dropped three, and received a grateful smile that nearly ruined his criminal posture. “Cake?” the woman asked. “Absolutely,” Bill said. “We’re here to steal presents,” Molly hissed. “Criminals need carbohydrates.” They moved toward the curtain, but a small boy blocked their path. He wore a cardboard crown and held a plastic sword. “Are you the surprise thieves?” he demanded. Molly froze. Bill crouched beside him. “No, young sir. We are highly trained birthday consultants.” “What do you consult?” “Whether the cake has enough frosting.” The boy considered this. “It doesn’t.” Bill nodded gravely. “A national emergency.” Behind them, the magician shouted, “And now, for my greatest trick!” The lights dimmed. Molly grabbed Bill’s sleeve, and together they darted behind the curtain. The presents filled a folding table: bright paper, curling ribbons, enormous bows. Molly reached for the largest box, but Bill intercepted her hand. “Wait,” he whispered. “What if these are all socks?” “Then we steal them with dignity.” A rustle came from beneath the table. The rabbit emerged, dropped the tiny vest, and glared at them. Bill slowly placed the box down. “I believe we’ve been outnumbered.” Molly looked toward the children, who were applauding the magician’s vanishing cake. She sighed, then picked up the rabbit’s vest. “We take one thing,” she said. “For evidence.” “And return it?” “Eventually.” They crept back into the party, carrying the vest like priceless contraband. Bill paused at the cake table, cut two slices, and handed one to Molly. “Retirement?” he asked. “After frosting,” she said. “Every great heist deserves a sequel.”
created_at
"2026-07-09T17:51:51.188Z"
7/9/2026, 5:51:51 PM
model
"openai/gpt-5.6-luna-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2372
moderation_latency
(null)
generation_time
6154
tokens_prompt
77
tokens_completion
1224
native_tokens_prompt
66
native_tokens_completion
999
native_tokens_completion_images
0
native_tokens_reasoning
273
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
0.00606
router
(null)
provider_responses
0
endpoint_id
"eebea444-34a5-4642-a766-cb319471d33a"
id
"resp_016e2f37e3c95f1d016a4fdfb7468081969e4e551a601e0dc5"
is_byok
false
latency
234
model_permaslug
"openai/gpt-5.6-luna-20260709"
provider_name
"OpenAI"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.18.0; linux; x64))"
http_referer
(null)
request_id
"req-1783619511-H75Jnj0IduiEMw5efTrK"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783619511-8AU5aIloMcluKyHV0OIN"
upstream_id
"resp_016e2f37e3c95f1d016a4fdfb7468081969e4e551a601e0dc5"
total_cost
0.00606
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
27.17%
Matches word count
n/a
neededClean
false
words
519
0.00%
Dialogue to Total Word Ratio
Ratio: 31.61%, Deviation: 18.39%
neededClean
false
wordsTotal
522
wordsDialogue
165
13.5829%