NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-09-01KX40EB13W94Q2R9JX0EG7CCA
Run Details
Model:
GPT-5.6 Sol (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
53.7s
Cost:
$0.137850
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
99.74%
Matches word count
(details)
98.28%
Dialogue to Total Word Ratio
(details)
99.0146%
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
At nine o’clock, Molly spread the birthday invitation across a crate of counterfeit bananas. Bill leaned over it, wearing spectacles with painted eyes on the lenses. The warehouse smelled of rope, dust, and somebody’s lunch. “Operation Cakewalk officially begins at noon. Thirty children, forty presents, and one deeply suspicious magician.” Bill tapped the map with a sausage. He had brought six for lunch and classified three as emergency equipment. Red arrows surrounded the community hall, though none pointed toward a door. Molly noticed this immediately. “Why do your arrows lead to the duck pond?” “Ducks improvise. Also, doors judge me.” Their disguises occupied a laundry basket: tiny party hats, two aprons, a papier-mâché dragon head, and shirts declaring them licensed balloon veterinarians. Bill selected the dragon head. It immediately swallowed his spectacles and sneezed glitter. “We need to look ordinary, Bill, not mythical.” “Children trust dragons; absolutely nobody trusts uncles.” Molly produced the schedule. At twelve, guests arrived. At twelve-ten, games began. At twelve-thirty, the magician released a rabbit. At twelve-thirty-one, according to Bill’s notes, civilization collapsed. The presents waited beside a cardboard castle downstairs in the echoing basement. “We slip in during musical statues, fill the banana crate, and quietly leave before cake.” Bill frowned so deeply the dragon frowned with him. He believed every plan required a diversion, preferably involving custard, gravity, or municipal brass bands. Unfortunately, their budget contained four coins, a button, and the sausages. “I’ll start a balloon emergency.” “You cannot start emergencies.” “Then I’ll improve an existing one.” They rehearsed beside the loading bay. Molly became a cheerful caterer, gliding with invisible sandwiches. Bill became a dragon suffering from confidential balloon trouble. On his third entrance, he collided with the crate and apologized. “Not to the crate.” “To whom, then?” “Nobody. Crates don’t have feelings.” “It looked hurt.” A problem emerged when Molly opened the crate. The false bottom, designed for rapid gift concealment, had become home to a gray cat. It stared at them with the cold authority of management reviewing expenses. “We cannot take a cat to parties.” “Why not? It’s already dressed better than me.” The cat climbed into the dragon head and refused further negotiation. Bill declared this an omen. Molly declared it a hygiene violation. Neither declaration moved the cat, which began purring like a badly parked motorcycle. “The cat joins us, but it gets no share.” “What about the rabbit?” “Union rules.” By midnight, the plan covered seven pages, three napkins, and the dragon’s nostril. Molly circled the exit. Bill drew a heroic portrait of himself carrying parcels while pursued by jealous toddlers. The cat edited it. “Remember: no improvising.” “No custard.” “Cruel, but fair.” “And no adopting the magician’s rabbit.” “Unreasonable.” At eleven fifty-nine the next morning, they stood across from the hall. Streamers fluttered above the entrance. Children shrieked inside. Bill lowered the dragon head, and the cat peered from its mouth like a tongue. “Ready?” “Ready. For wealth, glory, tiny wrapped boxes, and absolutely no sticky fingerprints on anything.”
created_at
"2026-07-09T17:57:37.45Z"
7/9/2026, 5:57:37 PM
model
"openai/gpt-5.6-sol-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5444
moderation_latency
(null)
generation_time
53336
tokens_prompt
77
tokens_completion
1270
native_tokens_prompt
66
native_tokens_completion
4584
native_tokens_completion_images
0
native_tokens_reasoning
3886
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
0.13785
router
(null)
provider_responses
0
endpoint_id
"a54c5de0-89bf-4ad7-a212-cf977eed918a"
id
"resp_0a200b69f625e0cd016a4fe111838c81948956974117b1665e"
is_byok
false
latency
227
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-1783619857-Qfhgxj4HkJBf6TlP8uZ6"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783619857-6Uu3CdK77wIDMnjrzt77"
upstream_id
"resp_0a200b69f625e0cd016a4fe111838c81948956974117b1665e"
total_cost
0.13785
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
99.74%
Matches word count
n/a
neededClean
false
words
504
98.28%
Dialogue to Total Word Ratio
Ratio: 29.41%, Deviation: 0.59%
neededClean
false
wordsTotal
510
wordsDialogue
150
99.0146%