NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHK97P1GCTXZ1B23KVSNGEXZ
Run Details
Model:
GPT-5
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
1m 49s
Cost:
$0.062592
Provider:
OpenAI
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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
They crouched behind a hedge shaped like a dinosaur that judged them with topiary disdain. Balloons bobbed above the community hall, each one a taunting thought bubble that read: Not yours. Molly adjusted a sequined beanie and squinted at the gift table visible through the window glare. Bill, armed with a clipboard that conferred dubious authority, practiced his most harmless smile until it creaked. A paper banner shouted HAPPY BIRTHDAY in letters so loud even the ants wore party hats. "Masks on, Bill; think sprinkles." "Sprinkles aren't stealth, Molly; confetti?" They had cased weddings, art shows, and once a beekeeper's picnic, but this, this was chaos with frosting. Behind the glass, pint-sized gladiators circled a cake like a sacred relic, guardians sworn by sugar. Parents orbited in exhausted ellipses, their coffee cups describing small, shaky moons. "Clown shoes distract witnesses; priceless." Bill checked the schedule he had entirely invented; it listed Cake O'Clock and a ceremonial piƱata weep. The plan was simple the way a Rube Goldberg daydream is simple: arrive, cart every present away, exit, pretend nothing happened. "Fine; we improvise, grab bows." Molly tugged a ribbon from her pocket, then another, until she looked less criminal than tumbleweed.
created_at
"2026-02-16T13:13:20.843801+00:00"
2/16/2026, 1:13:20 PM
model
"openai/gpt-5-2025-08-07"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
10764
moderation_latency
(null)
generation_time
109039
tokens_prompt
77
tokens_completion
1425
native_tokens_prompt
66
native_tokens_completion
6251
native_tokens_completion_images
0
native_tokens_reasoning
5952
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"completed"
usage
0.0625925
router
(null)
provider_responses
0
id
"resp_068710fb2b39c518016993178331bc8197853ad045fa8004a1"
status
200
is_byok
false
latency
147
endpoint_id
"7c2f859a-7890-4e8e-b1de-1cd1c0a800b4"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5-2025-08-07"
api_type
"completions"
id
"gen-1771247491-OHAwmmqZd8rwiTnzd9et"
upstream_id
"resp_068710fb2b39c518016993178331bc8197853ad045fa8004a1"
total_cost
0.0625925
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
99.9990%
Matches word count
n/a
neededClean
false
words
201
100.0000%
Dialogue to Total Word Ratio
Ratio: 9.95%, Deviation: 0.05%
neededClean
false
wordsTotal
201
wordsDialogue
20
99.9995%