NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-09-01KX40ZHYKJ1MH5EX9E8K5QWHZ
Run Details
Model:
GPT-5.6 Luna (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
28.5s
Cost:
$0.005232
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
99.98%
Matches word count
(details)
19.70%
Dialogue to Total Word Ratio
(details)
59.8414%
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 shaped like a regrettable rhinoceros, watching the brightly decorated house across the street. A paper banner reading HAPPY BIRTHDAY, OLIVER! sagged between two trees, while balloons bobbed from the porch like startled fruit. Through the windows, children sprinted in circles, parents carried trays of sandwiches, and one enormous magician appeared to be losing an argument with a rabbit. Bill adjusted his black woolly hat, which was too warm for June and had “World’s Best Uncle” stitched across the front. “Are you sure this is the place?” he whispered. Molly checked the address written on her wrist. “Number fourteen. Oliver’s party. Presents in the sitting room. Side gate unlocked.” “And the plan?” “We enter disguised as entertainers, distract the adults, collect the gifts, and leave through the conservatory.” Bill glanced at the house. “What are we disguised as?” “Entertainers.” “That’s not a disguise. That’s a career aspiration.” Molly ignored him and opened the suitcase between them. Inside lay two rubber noses, a deck of cards, three rolls of streamers, and a remote-controlled toy tank. Bill lifted the tank. “You said we were bringing professional equipment.” “It has smoke pellets.” “It has a duck sticker on it.” “The duck is tactical.” A shriek came from the garden. The magician had apparently produced a live chicken instead of the rabbit. Several children cheered. The chicken did not. Molly studied the party through a pair of opera glasses. “The presents are stacked beneath the window. Twenty-three boxes, two gift bags, and something wrapped in a blanket.” “That’ll be the pony,” Bill said. “Don’t be ridiculous. Nobody gives a pony to a six-year-old.” A miniature pony trotted past the window wearing a paper crown. Bill nodded gravely. “We’ll need a larger getaway vehicle.” Their original vehicle, a dented ice-cream van borrowed from a cousin who asked no questions and gave no explanations, waited around the corner. Unfortunately, its painted slogan—LICK YOUR WAY TO HAPPINESS—made stealth difficult. Molly had covered the words with a bedsheet, but the sheet had blown loose, revealing the word LICK in enormous blue letters. “We can’t carry everything,” Bill said. “We don’t have to carry everything. We only need to look as though we belong.” Molly produced two name badges. Hers read Professor M. Moonbeam. Bill’s read Dave. “Why am I Dave?” “Children trust Daves.” “Do they trust Daves with sacks of presents?” “Not if they ask questions. That’s why you’ll be a mime.” Bill stared at her. “A mime?” “A silent mime.” “I object to both parts.” They slipped through the side gate as the adults applauded the magician’s emergency escape from a laundry basket. Molly rolled the toy tank beneath the hedge and pressed the button. It trundled toward the house, coughing theatrical smoke. Every child in the garden screamed with delight and charged after it. Bill grinned. “The tactical duck has engaged.” Molly sprinted for the window. “Get the presents!” Behind them, the pony followed, dragging the magician’s trousers toward freedom.
created_at
"2026-07-09T18:07:01.595Z"
7/9/2026, 6:07:01 PM
model
"openai/gpt-5.6-luna-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2228
moderation_latency
(null)
generation_time
28383
tokens_prompt
77
tokens_completion
1016
native_tokens_prompt
66
native_tokens_completion
861
native_tokens_completion_images
0
native_tokens_reasoning
184
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
0.005232
router
(null)
provider_responses
0
endpoint_id
"eebea444-34a5-4642-a766-cb319471d33a"
id
"resp_05930c2ba207dd5c016a4fe345afe0819783e2a9d257547790"
is_byok
false
latency
228
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-1783620421-XdSjDDVQwkFhYsPuLOx7"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783620421-pUHQqGqgTZqGwKXwAZj8"
upstream_id
"resp_05930c2ba207dd5c016a4fe345afe0819783e2a9d257547790"
total_cost
0.005232
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
99.98%
Matches word count
n/a
neededClean
false
words
502
19.70%
Dialogue to Total Word Ratio
Ratio: 35.70%, Deviation: 5.70%
neededClean
false
wordsTotal
507
wordsDialogue
181
59.8414%