NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HVS7QEKAVJV4FA8X9SNVS
Run Details
Model:
GPT-6 Sol (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
17.4s
Cost:
$0.008292
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
99.74%
Matches word count
(details)
73.61%
Dialogue to Total Word Ratio
(details)
86.6789%
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 house, a garden, and what appeared to be a furious orange spider. “That’s the piñata,” she told Bill. “I thought it was the homeowner.” “The homeowner is a dentist. This is the party. Twelve children, one enormous cake, and, by my estimate, twenty-seven presents.” Bill studied the drawing with the grave attention of a man examining blueprints rather than a placemat. He had ordered a milkshake to make their meeting look innocent. Unfortunately, he was wearing a black turtleneck, black gloves, and a name badge that said BILL, CRIMINAL. “I found the badge in a costume shop,” he said. “I didn’t ask.” “You’ve been staring at it.” “I’m staring at the price tag still attached to your neck.” Bill tucked the tag into his collar. Molly folded her hands over the drawing before a passing waiter could see it. The waiter set down two bowls of ice cream with a smile. Neither thief had ordered ice cream, but both began eating immediately, since sending it back would attract attention. Outside, balloons bobbed above the dentist’s garden wall. Molly had spent all morning watching guests arrive. Every child had carried a gift bag or a box wrapped in paper so bright it could probably be seen from space. Now those boxes sat together on a table by the back door. “We go in as entertainers,” Molly said. “You’re a magician.” “I know one trick.” “You know half a trick. The coin never comes back.” “That’s why I need the money from the presents.” “You’ll do the trick, then. Keep everyone looking for the coin while I collect the gifts.” Bill considered this. He reached behind Molly’s ear and produced a damp napkin. “See? Already improving.” “That was tucked behind my ear.” “Then you’re improving too.” Molly pushed her bowl aside. It was nearly empty. She had intended to save room for any unattended cake, but planning a robbery had made her hungry. Across the street, the piñata swung in the breeze. The orange spider had eight lopsided legs and an expression of deep professional disappointment. A small girl appeared at the garden gate and waved at them. “She’s spotted us,” Bill whispered. “She’s waving. Wave back.” They waved. The girl crossed the street with the brisk purpose of someone old enough to delegate but too young to drive. “Are you the magicians?” she asked. Bill straightened. Molly smiled the smile she used when returning faulty electronics. “We are,” Molly said. “Good. The other one canceled. Come on. Dad says there’s cake for you afterward.” The girl marched back toward the gate. Bill looked at Molly. “What’s the plan now?” “The plan,” Molly said, rising, “is to take the job.” “And the presents?” “We’ll see how the trick goes.” Bill followed her across the street, practicing with the napkin. By the time they reached the garden, he had made it disappear into his sleeve twice. Molly suspected the coin would never stand a chance.
created_at
"2026-09-23T16:33:58.015Z"
9/23/2026, 4:33:58 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5848
moderation_latency
(null)
generation_time
17400
tokens_prompt
75
tokens_completion
959
native_tokens_prompt
66
native_tokens_completion
816
native_tokens_completion_images
0
native_tokens_reasoning
172
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
0.008292
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181238-yv3rmizGxij2PG1Mmm5w"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181238-N8jBV6qZBrLCmXU3l8qh"
upstream_id
"resp_0318cb76f6bc549d016ab3ff761f1c87d19a1fcd434a902981"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_0318cb76f6bc549d016ab3ff761f1c87d19a1fcd434a902981"
is_byok
false
latency
996
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.008292
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
99.74%
Matches word count
n/a
neededClean
false
words
504
73.61%
Dialogue to Total Word Ratio
Ratio: 32.48%, Deviation: 2.48%
neededClean
false
wordsTotal
505
wordsDialogue
164
86.6789%