NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37J0CM6DF01TPE6NK9S3NHQ
Run Details
Model:
GPT-6 Sol
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
16.1s
Cost:
$0.006852
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
90.48%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
45.2421%
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. It showed a square house, a triangular roof, and seventeen stick figures, each clutching what looked like a tiny refrigerator. “Those are presents,” she told Bill. “I assumed they were extremely successful children.” “Birthday party. Saturday, two o’clock. Front room for games, back room for cake. Presents on the dining table.” Bill studied the drawing. “Who made this?” “My nephew. He was promised five dollars and a balloon.” “You paid for intelligence with a balloon?” “The balloon was for me.” Outside, rain stitched silver lines down the café window. Inside, Bill slid his mug aside and unrolled a second document: a napkin covered in arrows. “Operation Birthday Surprise,” he said. Molly frowned. “That sounds like we’re delivering a surprise.” “We are. A terrible one.” “We need a better name.” “Operation All the Presents?” “Too obvious.” “Operation Most of the Presents?” “Bill.” He tapped his napkin. “We enter dressed as entertainers. I’ve acquired a magician’s cape, a collapsible top hat, and three rubber chickens.” “Why three?” “One’s a decoy.” Molly took a slow breath. Their last job had ended with Bill trapped inside a vending machine after mistaking the returns slot for a service hatch. She had promised herself she’d work alone. Then she’d remembered he owned a van. “Can you actually do magic?” she asked. “I can produce a coin from behind someone’s ear.” “Can you make thirty-seven wrapped boxes disappear?” “That depends how large the ear is.” A waitress arrived with two slices of pie. Bill covered the napkin with his elbow until she left, although the words STEAL PRESENTS were visible above it in blue ink. Molly turned the crayon drawing around. “There’s a problem. My nephew says every adult gets a name tag. His mother checks people at the door.” “Fine. We’ll be Uncle Bill and Aunt Molly.” “She knows me.” “Then I’ll be Uncle Bill and you’ll be someone else.” “The magician’s assistant?” “The magician’s accountant.” Molly looked at him. “Children need financial literacy,” Bill said. At the next table, an actual child watched them over a glass of milk. Molly folded the drawing. The child waved. On his wrist was a bright paper bracelet reading HAPPY BIRTHDAY, LEO. Molly’s stomach sank. “That’s the birthday boy.” Bill turned. Leo smiled, milk moustache gleaming, and raised a toy walkie-talkie. A tinny voice crackled from it. “Agent Leo, have you found the balloon lady?” Leo pressed the button. “Yes, Mum. She’s with a man who has chickens.” Bill stared at the bag beside his chair. One rubber beak poked through the zipper. Molly stood, leaving her pie untouched. “Change of plan.” “Escape through the kitchen?” “Buy a present.” Bill glanced at Leo, who was still watching. “And the chickens?” “Wrap them. All three.” Leo grinned as Molly headed for the counter. Behind her, Bill checked the price on the pie, then quietly slid one rubber chicken beneath his cape.
created_at
"2026-09-23T16:36:28.949Z"
9/23/2026, 4:36:28 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3431
moderation_latency
(null)
generation_time
16022
tokens_prompt
75
tokens_completion
843
native_tokens_prompt
66
native_tokens_completion
672
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
0.006852
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181388-CsEi1SLUDcH26WR3XY1o"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181388-OFhmSxpyEg7W53hb6CID"
upstream_id
"resp_0dbfbd8abd05920a016ab4000d1c3887d1ac3303d2eb0477b3"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_0dbfbd8abd05920a016ab4000d1c3887d1ac3303d2eb0477b3"
is_byok
false
latency
1463
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.006852
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
90.48%
Matches word count
n/a
neededClean
false
words
490
0.00%
Dialogue to Total Word Ratio
Ratio: 45.73%, Deviation: 15.73%
neededClean
false
wordsTotal
492
wordsDialogue
225
45.2421%