NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHKAEB5QR07DHQJD3XP4WY7R
Run Details
Model:
GPT-5 Nano
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
2m 30s
Cost:
$0.006238
Provider:
OpenAI
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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: Bill, the party's tonight; the presents deserve a vacation. Molly: Glide, yes, but softly; no sirens, no glittery alarms. Bill: We scout the gym closet, pretend it's balloon storage. Molly: We'll swap security keys with a kid's candy bracelet. Bill: Bracelet that unlocks chaos and confetti. Molly: Confetti is loud; we need silence, like sneaking cats. Bill: Then wear earmuffs; pretend we're part of the marching band. Molly: I brought a map drawn in frosting on napkins. Molly: We grab the bears, the scooters, and the glitter. Bill: No, we steal the presents, not their squeaky mascots. Molly: Right, the presents; wrapping paper like a treasure map. Bill: We'll stage a cleanup crew; distraction is our friend. Molly: You mean we pretend to sweep candy canes into a cart. Bill: Meanwhile, I stash the stuff behind a magician's wand. Molly: Isn't stealing from kids a little, you know, heartless? Bill: Heartless is our last resort after the cake orchestra. Molly: Anyway, we meet at midnight, discuss alibis, share jokes. Bill: Deal; bring the glitter wand, and we'll vanish, giggling. Outside, the party lights glitter as Molly and Bill whisper triumph, imagining stuffed penguins and overdue sleigh bells for fun.
created_at
"2026-02-16T13:35:08.984953+00:00"
2/16/2026, 1:35:08 PM
model
"openai/gpt-5-nano-2025-08-07"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
10313
moderation_latency
(null)
generation_time
150114
tokens_prompt
77
tokens_completion
1366
native_tokens_prompt
66
native_tokens_completion
15586
native_tokens_completion_images
0
native_tokens_reasoning
15232
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
0.0062377
router
(null)
provider_responses
0
id
"resp_00d9005021def0950169931c7608c88197af69da4fcfce6a50"
status
200
is_byok
false
latency
129
endpoint_id
"50329d77-04e1-4979-a184-c33030289476"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5-nano-2025-08-07"
api_type
"completions"
id
"gen-1771248757-N9EiycTFjRKs2rsIvgxS"
upstream_id
"resp_00d9005021def0950169931c7608c88197af69da4fcfce6a50"
total_cost
0.0062377
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Matches word count
n/a
neededClean
false
words
200
0.0000%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 90.00%
neededClean
false
wordsTotal
200
wordsDialogue
0
50.0000%