NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-09-01KX418SEBZMQ2N3Q175VW1H2Z
Run Details
Model:
GPT-5.6 Sol (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
19.6s
Cost:
$0.030270
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
20.19%
Matches word count
(details)
5.89%
Dialogue to Total Word Ratio
(details)
13.0400%
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 the birthday invitation across the café table, pinning its corners with a sugar bowl, a ketchup bottle, and Bill’s left elbow. A cartoon dragon grinned beneath the words PRINCESS POPPY’S SEVENTH BIRTHDAY. Glitter had migrated from the card onto Molly’s lockpicks. “The balloons are surveillance, Bill. Every one has a face, and every face is judging us.” “They’re helium, Molly. Their operational range is the ceiling.” “That’s exactly where nobody looks.” Bill studied the floor plan she had drawn on a napkin. It showed the house, garden, conservatory, and what appeared to be an octopus attacking a shed. “Is that the security system?” “That is a bouncy castle. Same principle, but louder.” Their target was circled in red: thirty-two presents stacked beside the birthday throne. According to Molly’s intelligence, the haul included dolls, robots, chemistry sets, three remote-control cars, and something called a Unicorn Spa Palace Deluxe. Bill sipped his tea. A strand of silver tinsel clung to his moustache, compromising his usual air of criminal dignity. “We enter disguised as entertainers,” Molly said. “You’ll be a magician. I’ll be your glamorous assistant.” “I don’t know any magic.” “You made twelve thousand pounds disappear in Brighton.” “That was accounting.” “Children won’t care. Pull something from a hat.” “What?” Molly glanced around the café. “A spoon. Children have low standards.” At the neighboring table, a toddler dropped a spoon and applauded. Molly pointed triumphantly. Bill examined the schedule. At two o’clock came musical statues; at two fifteen, cake; at two thirty, a performer named Captain Giggles. Molly had drawn a skull beside his name. “What happens to Captain Giggles?” “He receives an anonymous message canceling the booking.” “You wrote, ‘Come alone if you want your nose back.’” “It’s a clown-specific idiom.” Bill sighed. Outside, rain polished the street while a delivery van unloaded party supplies. Molly watched boxes of paper crowns go by with the concentration of a general observing artillery. Their escape route ran through the kitchen, over the garden wall, and into a waiting ice-cream van. Bill had objected that neither of them owned an ice-cream van, so Molly had added STEAL VAN FIRST to the plan. “Thirty-two presents,” Bill murmured. “How do we carry them past twenty children?” “We don’t. We announce a game called Pass the Parcel to the Van.” “That is offensively simple.” “Great crimes often are.” “What if the parents intervene?” “Face painting. Nobody argues while you’re holding a wet tiger.” Bill folded the napkin carefully. “And if the children chase us?” Molly’s confidence flickered. Children were fast, sticky, and unburdened by property law. She reconsidered the bouncy castle, the judging balloons, and Captain Giggles, who might possess clown allies. “We scatter sweets behind us.” “Bribery?” “Countermeasures.” Bill nodded slowly. It was the best bad plan they had devised all week. The toddler applauded again. Molly smiled. “See? We already have public support.”
created_at
"2026-07-09T18:12:04.18Z"
7/9/2026, 6:12:04 PM
model
"openai/gpt-5.6-sol-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4530
moderation_latency
(null)
generation_time
19516
tokens_prompt
77
tokens_completion
1126
native_tokens_prompt
66
native_tokens_completion
998
native_tokens_completion_images
0
native_tokens_reasoning
328
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
0.03027
router
(null)
provider_responses
0
endpoint_id
"a54c5de0-89bf-4ad7-a212-cf977eed918a"
id
"resp_0ee218f48490374e016a4fe47449ec8195b6d4788d6ae62326"
is_byok
false
latency
264
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-1783620724-dhpflLxjzc2epgZyaiZL"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783620724-GrUAIrDhCfFD8kgVRaVF"
upstream_id
"resp_0ee218f48490374e016a4fe47449ec8195b6d4788d6ae62326"
total_cost
0.03027
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
20.19%
Matches word count
n/a
neededClean
false
words
480
5.89%
Dialogue to Total Word Ratio
Ratio: 37.53%, Deviation: 7.53%
neededClean
false
wordsTotal
485
wordsDialogue
182
13.0400%