NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-09-01KX415FQCW48YQEVF9HJWYW5G
Run Details
Model:
GPT-5.6 Sol (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
19.8s
Cost:
$0.028710
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.21%
Matches word count
(details)
6.46%
Dialogue to Total Word Ratio
(details)
3.3393%
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 square house, a lopsided garden, and something labeled DANGER DRAGON. Bill squinted. “Is that the target?” “That’s a hydrangea.” “It has teeth.” “I’m bad at flowers. Focus. At three o’clock, twenty children enter through the front door. At three fifteen, cake. At three thirty, presents.” “And at three thirty-one, heartbreak.” “Don’t get sentimental. Those presents are worth thousands.” “They’re mostly wooden educational toys.” “Exactly. Have you seen what middle-class parents pay for a xylophone shaped like a sustainable whale?” Bill tapped the map. “How do we get inside?” “Disguises.” “Please say magicians.” “Princesses.” Bill recoiled. “I haven’t got the ankles.” “You’ll be a pirate.” “Pirates and princesses? Bit thematically confused.” “The birthday girl requested ‘everything.’ There’ll also be dinosaurs, astronauts, fairies, and one exhausted man making balloon animals.” “What are you going as?” Molly lifted a glittering pair of wings from her bag. “A burglar with moth problems?” “The Sugar Plum Fairy.” “You look more like the Tax Audit Fairy.” “That’s why children obey me.” Bill studied the schedule. “Where are the presents?” “Conservatory. Guarded by Grandma.” “Armed?” “Knitting needles.” “Professional?” “Forty years’ experience. Possibly Korean War.” “That was before forty years ago.” “She may have knitted through it.” A waitress passed. Molly folded the map beneath a menu. Bill whispered, “What’s the escape route?” “Kitchen, pantry, side gate, van.” “We have a van?” “We have a vehicle.” “The ice-cream tricycle is not a vehicle. It’s humiliation with pedals.” “It was cheap.” “It plays Greensleeves whenever we turn left.” “Then we turn right.” “For the entire getaway?” “We’ll escape in a spiral.” Bill sighed. “Suppose a child sees us?” “You distract them.” “How?” “Pirate voice. Treasure hunt. Existential discussion about parrots. Improvise.” “I don’t improvise well under pressure.” “You told a police officer you were transporting an emotional wardrobe.” “It was visibly distressed.” “And if Grandma challenges us, I deploy Operation Cupcake.” “Which is?” “You offer her one.” “That’s not an operation.” “It is if the cupcake contains brandy.” Bill lowered his voice. “Molly, are we really stealing birthday presents from children?” “We’re thieves.” “We have standards.” “You stole a vicar’s lawnmower.” “He had two.” “These children have too much stuff. They’ll barely notice.” “They’ll notice the empty table.” “We leave replacements.” Molly produced a sack of potatoes, each decorated with googly eyes. Bill stared. “You made potato dolls?” “Handcrafted collectibles.” “This one says Kevin.” “They all say Kevin.” A long silence settled between them. Bill picked up a potato. “This is the saddest face I’ve ever seen.” “That’s because you’re holding it upside down.” He turned Kevin over. “Worse,” he said. Molly gathered the map and wings. “Three o’clock. Pirate outfit. Bring the tricycle.” Bill tucked Kevin into his coat. “Fine, but we take only expensive presents.” “Agreed.” “And no bicycles.” “Agreed.” “And if a child cries, we abort.” “Agreed.” “And I’m not wearing eyeliner.” Molly smiled. “Then nobody will believe you’re a pirate.” Bill groaned. “Do I at least get a sword?” “Foam.” “Hook?” “Plastic.” “Parrot?” “Kevin.” From inside Bill’s coat, the potato’s googly eyes rattled. Bill nodded solemnly. “Then the crew is assembled.”
created_at
"2026-07-09T18:10:15.927Z"
7/9/2026, 6:10:15 PM
model
"openai/gpt-5.6-sol-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3049
moderation_latency
(null)
generation_time
19663
tokens_prompt
77
tokens_completion
1290
native_tokens_prompt
66
native_tokens_completion
946
native_tokens_completion_images
0
native_tokens_reasoning
138
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
0.02871
router
(null)
provider_responses
0
endpoint_id
"a54c5de0-89bf-4ad7-a212-cf977eed918a"
id
"resp_07e82864e681b025016a4fe40802a881959dfa0683dcc643b1"
is_byok
false
latency
199
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-1783620615-c6YVcB4dupon4rmfOeqD"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783620615-KCu0OjymetoPVFvb5B3O"
upstream_id
"resp_07e82864e681b025016a4fe40802a881959dfa0683dcc643b1"
total_cost
0.02871
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
0.21%
Matches word count
n/a
neededClean
false
words
528
6.46%
Dialogue to Total Word Ratio
Ratio: 77.40%, Deviation: 7.40%
neededClean
false
wordsTotal
531
wordsDialogue
411
3.3393%