NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-09-01KX40E3Z3GT9EPXZ2X7BYKATX
Run Details
Model:
GPT-5.6 Sol (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
19.7s
Cost:
$0.028110
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
60.28%
Matches word count
(details)
18.57%
Dialogue to Total Word Ratio
(details)
39.4244%
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 lowered her binoculars and frowned at the balloon-covered community hall. “Security is tighter than expected.” Bill peered through the windshield. “That’s a golden retriever wearing a paper crown.” “Exactly. Nobody suspects the dog.” “I suspect the dog.” “Then he’s already inside your head.” The retriever wagged at a passing butterfly. Bill opened a packet of crisps. “Review the plan.” “We infiltrate the party disguised as entertainers, locate the presents, load them into the van, and vanish before anyone notices.” “We’re disguised as a horse.” “A pantomime horse.” “A pantomime horse with one leg missing.” “The advertisement said ‘minor stitching required.’” “The advertisement lied. Also, why are we stealing children’s presents?” “Because, Bill, rich parents buy excellent presents.” “The invitation says Alfie is turning four.” “Four-year-olds get drones now.” “I got a wooden spoon.” “And look how you turned out.” Bill considered this, then offered her a crisp. “Who are we supposed to be entertaining?” “The children.” “How?” “Horse things.” “Name three horse things.” “Galloping, neighing, and standing in fields looking emotionally unavailable.” “I can do the third one.” Molly pulled the costume from the back seat. Its faded purple head wore a crooked horn. “This is a unicorn.” “It’s aspirational.” “It smells like soup.” “That’s our cover scent.” They wrestled into the costume. Molly occupied the front while Bill stumbled behind, clutching her belt. “I can’t see.” “You’re the back end. History rarely demands vision from the back end.” “What if we encounter stairs?” “Believe in me.” “I believed in you during the museum job.” “And we escaped.” “We were never inside the museum.” “Which made escaping effortless.” They shuffled across the road and entered the hall. Children shrieked. Molly froze. “They recognize us.” “They’re delighted.” “No one has ever made that noise from delight.” A small boy punched the unicorn’s nose. Another attached a balloon to Bill’s tail. “Maintain character,” Molly hissed. “What’s my character?” “Cheerful hindquarters.” “I trained at drama school for six weeks.” “Then use your craft.” They pranced toward a mountain of wrapped parcels beside the cake table. Bill kicked over a bowl of jelly. “Target located.” “Excellent. Begin extraction.” “With what hands?” Silence filled the costume. “You were bringing the sack.” “You said you were bringing it.” “I brought the binoculars.” “Very useful indoors.” The crowned retriever trotted over, seized the unicorn’s tail, and pulled. The costume split neatly down the middle. Molly toppled into the presents. Bill landed in the jelly. A woman applauded. The children cheered wildly. Bill lifted his dripping head. “They think it’s part of the act.” Molly emerged wearing three gift bows and holding a toy fire engine. “New plan.” “Run?” “Take a bow, eat cake, and leave with our dignity.” “Can we steal the fire engine?” “No.” “The wooden spoon of my childhood is weeping.” “Fine. But only if you promise not to drive it.” “It has pedals.” “Then the getaway is saved.”
created_at
"2026-07-09T17:57:30.229Z"
7/9/2026, 5:57:30 PM
model
"openai/gpt-5.6-sol-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4706
moderation_latency
(null)
generation_time
19598
tokens_prompt
77
tokens_completion
1103
native_tokens_prompt
66
native_tokens_completion
926
native_tokens_completion_images
0
native_tokens_reasoning
249
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
0.02811
router
(null)
provider_responses
0
endpoint_id
"a54c5de0-89bf-4ad7-a212-cf977eed918a"
id
"resp_001e6375adca7917016a4fe10a5c448196b8ee8cb0847a4250"
is_byok
false
latency
469
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-1783619850-ikOIeyWzZVhH0KHObA97"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783619850-4nYro0nXV17Nmx8K7bjN"
upstream_id
"resp_001e6375adca7917016a4fe10a5c448196b8ee8cb0847a4250"
total_cost
0.02811
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
60.28%
Matches word count
n/a
neededClean
false
words
485
18.57%
Dialogue to Total Word Ratio
Ratio: 64.20%, Deviation: 5.80%
neededClean
false
wordsTotal
486
wordsDialogue
312
39.4244%