NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HXFYA2KPH3PEDH3G5STQP
Run Details
Model:
GPT-6 Sol (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
17.8s
Cost:
$0.008602
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
57.38%
Dialogue to Total Word Ratio
(details)
28.6877%
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 map across the café table. It showed a house, a garden, and forty-seven balloons, each drawn with the care of a hostage sketch. “Where did you get this?” “Birthday boy’s invitation.” “You stole an invitation from a child?” “I borrowed it from a recycling bin. Keep up, Bill.” Bill squinted at the map. “Why is the garage labeled ‘dragon’?” “Because he’s six. That’s where the presents are. His mother posted a picture of the party setup.” “All the presents?” “All of them.” “That’s ambitious. Last time you said ‘all,’ we left with one shoe and a decorative gourd.” “The shoe was Italian.” “It was for a doll.” Molly tapped the map. “We go in at two, during the magician’s act. You’re dressed as a magician.” “What’s the actual magician dressed as?” “That’s his problem.” “He’ll notice another magician.” “Then you tell him you’re the regional manager.” “Of magic?” “Obviously.” Bill stirred his coffee. “And you?” “I’m the balloon safety inspector.” “That’s not a job.” “Neither is regional manager of magic. Yet here we are.” Outside, a bus hissed to a stop. Molly folded the map before anyone could see the dragon. “Suppose they ask for credentials.” “I made a badge.” “What does it say?” “BALLOON SAFETY.” “That’s not a credential. That’s two nouns having an argument.” “Three words. You’re counting badly, which worries me because you’re in charge of the presents.” “I thought you were in charge of the presents.” “I’m in charge of distracting the parents. You put the gifts in the big blue sack.” “What big blue sack?” “The one you brought.” Bill looked under the table. “I brought a lunch bag.” “How many presents fit in it?” “Depends. Are they sandwiches?” Molly leaned back. “Fine. We use the birthday boy’s gift tablecloth. Bundle everything inside it.” “Won’t he notice his presents disappearing?” “He’ll be watching the magician.” “Me?” “The real one. You’ll be in the garage.” “What am I doing there?” “Opening the side door so I can carry the bundle out.” Bill frowned. “I thought you were distracting the parents.” “I can do both. Watch.” She waved brightly at a passing waiter while sliding Bill’s teaspoon into her pocket. Bill watched the spoon vanish. “That’s impressive.” “Thank you.” “Give it back.” “After the job.” A silence settled between them. Bill studied the crayon drawing again. “There’s a note on the back,” he said. “‘Please bring one wrapped gift for the birthday charity drive. No presents for Oliver.’” Molly held out her hand. Bill surrendered the invitation. “So,” he said, “we’d be stealing gifts meant for charity.” “Yes.” “Does that change the plan?” Molly examined the cheerful little dragon, then pushed his teaspoon across the table. “Yes. We’re stealing the magician’s job. I already bought the hat.”
created_at
"2026-09-23T16:34:54.034Z"
9/23/2026, 4:34:54 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
6246
moderation_latency
(null)
generation_time
17778
tokens_prompt
75
tokens_completion
968
native_tokens_prompt
66
native_tokens_completion
847
native_tokens_completion_images
0
native_tokens_reasoning
204
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
0.008602
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181294-6BEzlvUACmDXiGhD0dw6"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181294-eqIr7twerJhBlJXu2zVc"
upstream_id
"resp_0af7d9d892c13967016ab3ffae276087d1ae2908417aa89bc8"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_0af7d9d892c13967016ab3ffae276087d1ae2908417aa89bc8"
is_byok
false
latency
619
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.008602
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
464
57.38%
Dialogue to Total Word Ratio
Ratio: 73.33%, Deviation: 3.33%
neededClean
false
wordsTotal
465
wordsDialogue
341
28.6877%