NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-09-01KX41MY298DTHK1NCQAFVJQKS
Run Details
Model:
GPT-5.6 Sol
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
15.7s
Cost:
$0.021120
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
51.93%
Matches word count
(details)
14.00%
Dialogue to Total Word Ratio
(details)
32.9639%
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 a crayon map across the café table, pinning each corner with a sugar packet. The map depicted a mansion, a garden, and what appeared to be a furious purple horse. “Your reconnaissance is terrible,” Bill said. “It was drawn by our informant.” “Your informant is six.” “Six and a half. Show some respect.” Outside, rain streaked the windows. Inside, the lunch crowd chatted, spoons clinked, and two professional thieves contemplated the most dangerous target of their careers: Daisy Featherstone’s seventh birthday party. Molly tapped a blue rectangle. “This is the ballroom. Presents go here.” “How many?” “Forty, perhaps fifty. Dolls, games, gadgets. One box may contain a remote-controlled helicopter.” Bill frowned. “That explains the anti-aircraft gun.” “That’s a chocolate fountain.” “Purple crayon lacks nuance.” Molly moved a biscuit onto the map. It left crumbs over the ballroom. “At fourteen hundred hours, the magician begins. Every child, parent, and security guard will watch him pull a rabbit from a hat.” “What about the rabbit?” “Not our problem.” “Rabbits notice things.” “You’re frightened of rabbits.” “I respect their surveillance capabilities.” Their plan required costumes. Molly would enter as Princess Petunia, a beloved storybook royal with a wand, a tiara, and a contractual inability to frown. Bill would pose as her cheerful assistant, Mr. Sparkles. Bill stared at the sequined jacket hanging from Molly’s chair. “I won’t wear that.” “It has hidden pockets.” “I’ll wear that.” At precisely two o’clock, Princess Petunia swept through the mansion gates, waving regally. Mr. Sparkles followed, jingling with every furious step. A butler checked their names against a clipboard. “You’re early,” he said. “Royalty is never early,” Molly replied. “Everyone else is late.” The butler nodded, intimidated by protocol, and admitted them. The ballroom roared with sugar-powered children. Balloons crowded the ceiling. A magician flapped beside a smoking top hat. Along the wall stood a mountain of presents wrapped in silver, gold, and dinosaur paper. Bill whispered, “We’ll need three sacks.” “Two. Avoid anything shaped like a drum kit.” They sidled toward the gifts. Molly produced collapsible bags. Bill reached for the helicopter-sized box. A small hand seized his sleeve. “You’re not Mr. Sparkles,” Daisy said. Bill froze. “Of course I am.” “Mr. Sparkles has a moustache.” “It fell off.” “Why?” “Seasonal moulting.” Daisy narrowed her eyes. Behind her, thirty children abandoned the magician and advanced with the silent purpose of tiny detectives. Molly raised her wand. “Who wants a treasure hunt?” Every hand shot up. “Excellent. Find the two suspicious adults disguised as entertainers.” The children cheered and tackled Bill first. Molly made it three steps before a fairy princess caught her ankle. Ten minutes later, both thieves sat gift-wrapped beneath the chocolate fountain while Daisy’s mother called the police. Bill sighed through a silver bow. “The rabbit warned them.” Beside the magician, the rabbit stared smugly. Molly nodded. “Next time, we rob adults.” “Fewer biscuits.” “Worse security.”
created_at
"2026-07-09T18:18:42.146Z"
7/9/2026, 6:18:42 PM
model
"openai/gpt-5.6-sol-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
741
moderation_latency
(null)
generation_time
15382
tokens_prompt
77
tokens_completion
934
native_tokens_prompt
66
native_tokens_completion
693
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
0.02112
router
(null)
provider_responses
0
endpoint_id
"a54c5de0-89bf-4ad7-a212-cf977eed918a"
id
"resp_04dc1f8bdbe562f0016a4fe60236f08196bcb8ef0d46c4effb"
is_byok
false
latency
228
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-1783621122-NmkbnEkNwDYZQWxa40vl"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783621122-J1mbUs3JTIftS1qRY44k"
upstream_id
"resp_04dc1f8bdbe562f0016a4fe60236f08196bcb8ef0d46c4effb"
total_cost
0.02112
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
51.93%
Matches word count
n/a
neededClean
false
words
484
14.00%
Dialogue to Total Word Ratio
Ratio: 36.27%, Deviation: 6.27%
neededClean
false
wordsTotal
488
wordsDialogue
177
32.9639%