NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHKA2VTWSAWG7YTMXXXG8Q6Y
Run Details
Model:
GPT-5
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
1m 34s
Cost:
$0.049362
Provider:
OpenAI
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 50% 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
The community hall pulsed with cake-fueled chaos, a bouncy castle wheezing like an accomplice with asthma. Streamers drooped, balloons squeaked, and a piñata grinned with paper teeth over a barricade of glittering gifts. Molly mapped exits with a frosting-smeared brochure while Bill polished a red foam nose he insisted was tactical camouflage. Parents milled like suspicious guards; toddlers patrolled with juice boxes and sticky jurisdiction. Their plan hinged on timing, confidence, and the universal distraction of a conga line. When the DJ hit maximum whimsy, they would ghost in, bag everything cheerful, and vanish behind the magician's smoke puff. Perfect. Molly: Clown entry at ten hundred; smiles are our grapples. Bill: Copy that; honk twice if security notices parachute pants. Molly: I'll flank the cake; you tango-pivot toward the piñata. Bill: Distraction move alpha; I start an emergency nap-time drill. Molly: Locate the present mountain; prioritize robots, puzzles, anything noisy. Bill: Bag structure confirmed; glitter-proof, crumb-resistant, deceptively festive, also monogrammed. Molly: Watch the dinosaur balloon; its tail decapitated my dignity. Bill: If cornered, deploy confetti; nobody arrests joy while sneezing. Molly: Exit through the magician; pretend we're volunteers returning doves. Bill: And if asked names, we're Cake Logistics, Sprinkles West.
created_at
"2026-02-16T13:27:55.825907+00:00"
2/16/2026, 1:27:55 PM
model
"openai/gpt-5-2025-08-07"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
9752
moderation_latency
(null)
generation_time
93369
tokens_prompt
77
tokens_completion
1392
native_tokens_prompt
66
native_tokens_completion
4928
native_tokens_completion_images
0
native_tokens_reasoning
4608
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
0.0493625
router
(null)
provider_responses
0
id
"resp_00f4d0765a4dc9b90169931afdde988197bdc478c576072896"
status
200
is_byok
false
latency
155
endpoint_id
"7c2f859a-7890-4e8e-b1de-1cd1c0a800b4"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5-2025-08-07"
api_type
"completions"
id
"gen-1771248381-JCBquDhfaOkIZXFgbnyf"
upstream_id
"resp_00f4d0765a4dc9b90169931afdde988197bdc478c576072896"
total_cost
0.0493625
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
98.7124%
Matches word count
n/a
neededClean
false
words
206
0.0000%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 50.00%
neededClean
false
wordsTotal
206
wordsDialogue
0
49.3562%