NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HTR4T3X9HRK71K57KSXX9
Run Details
Model:
GPT-6 Sol
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
17.1s
Cost:
$0.007682
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
45.06%
Dialogue to Total Word Ratio
(details)
22.5323%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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-drawn floor plan across the café table. A smiling giraffe labeled the room she’d circled in red. “Reconnaissance,” she said. “The party starts at two. At two fifteen, the magician arrives. At two thirty, everyone watches him pull a rabbit from a hat.” Bill studied the drawing. “Why is the kitchen labeled ‘lava’?” “Birthday boy’s instructions. He was very firm about it.” “And this purple thing?” “The present table.” “It says ‘treasure mountain.’” “That’s how I know he’s our mark.” Bill lowered his voice. “Molly, how many presents are we talking?” “Twenty-three on the invitation list. Maybe more if anyone brings a sibling.” “You’re counting on bonus presents?” “I’m accounting for variables.” The waiter set down two coffees and looked at the drawing. “Nice giraffe.” “Thank you,” Molly said. “He’s got six legs,” Bill muttered. “Five. One’s a tail.” The waiter went away looking unconvinced. Bill tapped the map. “How do we get into the party?” “Disguises. I’ve got a clown costume for you.” “I don’t do clowns.” “Fine. You can be the magician’s assistant.” “Does he have an assistant?” “He will when you show up holding his rabbit.” “We’re stealing the rabbit too?” “We’re borrowing it. He keeps it in a carrier by his car while he sets up.” Bill sat back. “You’ve put more thought into the rabbit than the presents.” “The rabbit gets us through the door. I’ll go in as a parent.” “Whose parent?” “I’ll find out when I arrive. Someone always wanders off to the loo.” “That doesn’t make you their parent.” “It gives me time to carry a present.” “We need twenty-three trips.” Molly pointed to a drawing of a rectangle beside the kitchen. “Wheelie bin. I’ve lined it with a clean tablecloth. We fill it, roll it out the side gate.” “Past the actual parents?” “They’ll be watching the magician.” “For twenty-three presents?” “He does a forty-minute show.” Bill finished his coffee. “What are we doing with them afterward?” “Opening them, obviously.” “I mean, why? Since when do we steal from children?” “Since this morning.” Molly took a folded sheet from her pocket. “I got this through my letterbox.” Bill read it. “‘Dear thief, I know you took the silver candlesticks from our house. Please put them back.’ Who wrote this?” “The birthday boy. We climbed through his window by mistake last Tuesday.” “He saw us?” “He watched from the stairs. Then he drew us on the back.” Bill turned the page. Two stick figures hauled candlesticks beneath a large moon. One had a surprisingly accurate beard. “He’s seven,” Molly said. “He’s already better at identifying suspects than the police.” “So we steal his birthday presents?” “No. We steal them, put the candlesticks in a box, and sneak everything back before cake. He can tell his parents he caught us.” Bill looked at her. “You could just return the candlesticks.” “And waste the wheelie bin?” Across the café, a balloon burst. Bill flinched, scattering sugar packets across the map. Molly gathered them before they could fall. “That’s another thing. You’ll need to stay calm.” “I am calm.” “You’ve crushed a packet in your fist.” Bill opened his hand. Sugar poured onto the giraffe. “Good,” Molly said. “Now it looks like a party.”
created_at
"2026-09-23T16:33:24.131Z"
9/23/2026, 4:33:24 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2492
moderation_latency
(null)
generation_time
17080
tokens_prompt
75
tokens_completion
951
native_tokens_prompt
66
native_tokens_completion
755
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
0.007682
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181204-ZN8sRxknfVsTdON5Nq1w"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181204-78yyarzoIPjRQ8XB33sD"
upstream_id
"resp_0f7f23c578c1494d016ab3ff543d4887d1be122259e9a7eb84"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_0f7f23c578c1494d016ab3ff543d4887d1be122259e9a7eb84"
is_byok
false
latency
740
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.007682
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
542
45.06%
Dialogue to Total Word Ratio
Ratio: 73.99%, Deviation: 3.99%
neededClean
false
wordsTotal
546
wordsDialogue
404
22.5323%