NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHK9ARYE1F9YGBRCJAS08JDX
Run Details
Model:
GPT-5
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
1m 14s
Cost:
$0.053582
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
Molly unfolded a crayon-smeared flyer advertising a backyard extravaganza and grinned. The plan wasn’t sleek; it wore cupcake icing and sticky fingerprints. Bill paced the alley’s chalk constellations, imagining sentries in paper hats and juice-box bribes backfiring. Somewhere, a bounce house sighed like a sleepy dragon. Confetti drifted from a previous surrender. They rehearsed smiles, compared camouflage—pastel hoodies, sneakers, an air of innocent bewilderment. Presents meant decoys, tissue paper traps, and tags weaponized by handwriting. Their getaway cart, a squeaky wagon, waited beneath a disgruntled piñata. Timing, they agreed, hinged on the birthday song’s note and the inevitable sugar crash. "Molly, this is low, even for us, you know. Right?" "Bill, relax; it's philanthropy inverted: we relieve future clutter, heroically." "Also, balloons squeak like alarms; we tiptoe, we succeed, yes?" "Tiptoe past toddlers? They detect crumbs at fifty yards, commander." "Good, then bribe the clown with coupons and existential reassurance." "Coupons expire, Molly; clowns value secrets, like where cake hides." "Perfect, whisper: the cake is square, the knives are sporks." "Sporks unnerve me; presents don't, except the ones that giggle." "Then avoid giggles; target boxes labeled From Aunt, safely sweatered." "And if caught, improvise a puppet show about sharing responsibly."
created_at
"2026-02-16T13:14:26.724758+00:00"
2/16/2026, 1:14:26 PM
model
"openai/gpt-5-2025-08-07"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
7241
moderation_latency
(null)
generation_time
73921
tokens_prompt
77
tokens_completion
1160
native_tokens_prompt
66
native_tokens_completion
5350
native_tokens_completion_images
0
native_tokens_reasoning
4992
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
0.0535825
router
(null)
provider_responses
0
id
"resp_0b9ca32ed7d97d9d01699317e877948197aad2efc8fba0f3da"
status
200
is_byok
false
latency
149
endpoint_id
"7c2f859a-7890-4e8e-b1de-1cd1c0a800b4"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5-2025-08-07"
api_type
"completions"
id
"gen-1771247592-GjdR6gBUeeGzrnPpOYb3"
upstream_id
"resp_0b9ca32ed7d97d9d01699317e877948197aad2efc8fba0f3da"
total_cost
0.0535825
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
99.9190%
Matches word count
n/a
neededClean
false
words
203
99.9970%
Dialogue to Total Word Ratio
Ratio: 49.26%, Deviation: 0.74%
neededClean
false
wordsTotal
203
wordsDialogue
100
99.9580%