NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HHXMJBH0KX74QSTTA15TQ
Run Details
Model:
GPT-6 Sol
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
14.6s
Cost:
$0.007272
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0008%
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 crayon drawing across the café table. A lopsided house stood beneath a purple sun; beside it, a rectangle was labeled PRESENTS in green capital letters. “Is that your floor plan?” Bill asked. “It was made by the birthday girl.” “Why does the garage have teeth?” “That’s a bouncy castle. Focus.” Bill peered closer. “How did you get it?” “She gave it to me. I’m her aunt.” “You have a niece?” “For this afternoon.” The café was full of parents drinking coffee with the stunned expressions of people who had recently paid for balloons. Molly had chosen it because no one would notice two adults discussing a children’s party over chocolate muffins. Bill had chosen the muffins. “The target is at four,” Molly said. “Presents go on the dining-room table. We arrive as entertainers, hold everyone’s attention, and move the goods.” Bill swallowed. “Who moves them?” “You do.” “What are you doing?” “Entertainment.” “You can’t entertain children.” “I’ve entertained you for years.” Bill gave this some thought, then took another bite. Molly had found the party through a neighborhood noticeboard. It advertised face painting, games, and a magician called The Amazing Alfredo. Alfredo had canceled after trapping his cape in a revolving door. Molly had replied from a newly created address, promising a replacement act with “years of experience delighting young audiences.” She had not specified what kind of years. “Magic, then?” Bill asked. “Magic and a disappearing-act finale.” “Can you do tricks?” “I can pull a coin from your ear.” “You put it there.” “That’s the foundation of most magic.” Bill looked at the drawing again. A stick figure in the garden held a red circle above its head. Underneath, in uneven letters, the artist had written DAD WITH BBQ. “Barbecue in November?” “British optimism. Now, your job is to bring a large prop trunk. When I ask the children to close their eyes, you put the presents inside.” “All of them?” “All of them.” “What if one’s too big?” “We leave the bicycle.” Bill sat back. “There’s a bicycle?” “Potentially.” “Can we take it?” “Bill.” He sighed. Molly slid a second sheet across the table: a timetable written on a napkin. Bill read it, frowned, and turned it upside down. “You’ve scheduled cake before the robbery.” “I’ve scheduled cake before your part of the robbery. No one watches the dining room when cake comes out.” “Everyone watches cake.” “Exactly.” Bill nodded slowly. A child at the next table leaned over and pointed at Molly’s drawing. “That’s my house,” the child said. Molly held very still. The girl had purple icing on her chin and a plastic tiara caught in her hair. Bill looked from the girl to the drawing. “Small world.” “Are you the new magician?” she asked Molly. “I am.” “Great. Mum says you can’t come now.” “Why not?” “Alfredo got out of the door.” The girl reclaimed her drawing and hurried back to her family. Molly watched her go. Bill watched Molly. “Plan B?” he asked. Molly picked up her muffin. “We eat these, then go home.” “I like Plan B.” Outside, a gust of wind rattled the café window. Molly took a bite and admitted, privately, that she did too.
created_at
"2026-09-23T16:28:34.842Z"
9/23/2026, 4:28:34 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2633
moderation_latency
(null)
generation_time
14561
tokens_prompt
75
tokens_completion
908
native_tokens_prompt
66
native_tokens_completion
714
native_tokens_completion_images
0
native_tokens_reasoning
0
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
0.007272
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790180914-Rt5pOcUMemVz48PXhrTp"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790180914-11X2QFKdGhpMzeu3O7JK"
upstream_id
"resp_073e91cfc76ea3d2016ab3fe32f03487d18e519d74971a7441"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_073e91cfc76ea3d2016ab3fe32f03487d18e519d74971a7441"
is_byok
false
latency
1291
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.007272
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
537
0.00%
Dialogue to Total Word Ratio
Ratio: 44.90%, Deviation: 14.90%
neededClean
false
wordsTotal
539
wordsDialogue
242
0.0008%