NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-09-01KX403P0BCCNQN7NBBNMME0AB
Run Details
Model:
GPT-5.6 Sol
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
8.7s
Cost:
$0.009450
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
81.27%
Matches word count
(details)
99.97%
Dialogue to Total Word Ratio
(details)
90.6211%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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 the crayon map across the dashboard. “Right, Bill. At fourteen hundred hours, the clown arrives.” “Armed?” “With balloon animals.” “Savage.” “The parents gather in the kitchen. The children begin musical chairs. That’s our window.” “We climb through it?” “No, Bill, our opportunity window.” “Good. I hate climbing. These trousers are tactical corduroy.” “You’ll enter disguised as a magician.” “I only know one trick.” “The disappearing sausage?” “It’s less impressive since the dog learned where I hide it.” “Focus. I’ll pose as the princess entertainer.” “You’ve got the beard for it.” “I’ll shave.” “You said that before Aunt Maureen’s wedding.” “That beard got us free cake.” “And a restraining order.” “The presents are stacked beside the bouncy castle. We wheel them out in the giant rabbit costume.” “Who wears the rabbit?” “You.” “I wore the goose last time.” “And you were magnificent.” “I got kicked by a toddler.” “You hissed at him.” “I was in character.” A horn honked outside. Both thieves froze. “What was that?” “The birthday clown.” “He’s early.” “Abort?” “Never. Initiate Plan B.” “What’s Plan B?” “We attend politely, eat jelly, and steal nothing.” “That’s not a heist.” “Bill, there’s a party bag for every guest.” “You criminal genius.” They adjusted their paper crowns and rang the bell.
created_at
"2026-07-09T17:51:48.244Z"
7/9/2026, 5:51:48 PM
model
"openai/gpt-5.6-sol-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
954
moderation_latency
(null)
generation_time
8583
tokens_prompt
77
tokens_completion
460
native_tokens_prompt
66
native_tokens_completion
304
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
0.00945
router
(null)
provider_responses
0
endpoint_id
"a54c5de0-89bf-4ad7-a212-cf977eed918a"
id
"resp_0474b58fe66d4508016a4fdfb4669081949709ccbaf8c171ac"
is_byok
false
latency
298
model_permaslug
"openai/gpt-5.6-sol-20260709"
provider_name
"OpenAI"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.18.0; linux; x64))"
http_referer
(null)
request_id
"req-1783619508-GXHbaQLUOIzOtKClj5l6"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783619508-HrEkzAm14GoeED4c7vPy"
upstream_id
"resp_0474b58fe66d4508016a4fdfb4669081949709ccbaf8c171ac"
total_cost
0.00945
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
81.27%
Matches word count
n/a
neededClean
false
words
212
99.97%
Dialogue to Total Word Ratio
Ratio: 88.68%, Deviation: 1.32%
neededClean
false
wordsTotal
212
wordsDialogue
188
90.6211%