NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHKAXEDWV27KG4WC8AY9CQQ5
Run Details
Model:
GPT-5 Mini
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
35.2s
Cost:
$0.007726
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: Checklist? Bill: Costume? Molly: Clown from the inflatable castle. You? Bill: Balloon-animal distraction. Also I learned a magic trick. Molly: Good. Gifts are shiny, wrapped, and suspiciously sugar-fueled. Bill: Kids are fast. Cake is slower. Parents are blind. Molly: We only take off-brand action figures and glitterless jewelry. Bill: And chocolate? Molly: Only emergency chocolate. No molasses disasters. Bill: Escape route? Molly: Balloon arch to the alley, scooter stash behind the piñata vendor. Bill: Piñata vendor? We're friends with the piñata vendor? Molly: We bribe with balloon hats. Bill: Timing? Molly: When the magician sawing himself—no, wait—the magician sawing the assistant. Actually when everyone watches the bouncy castle. Bill: Signal? Molly: My cough. Your honk? Bill: Honk will sound like a kazoo. Classic. Molly: Remember: tiny shoes, big pockets. Bill: Also return note: Thanks for the toys. Keep the taxes. Molly: Comedy, not chaos. Bill: Ready? Molly: Ready. Bill: Steal the thunder and the cookies. Molly: Don't forget the wrapping paper recycler. Bill: We recycle? It's a heist, not a landfill. Molly: Fine. Take the gift bags too. Bill: And the goodie bags. Small, but ceremonial. Molly: Meet at midnight on the playground merry-go-round. Bill: Gumboots for stealth puddle.
created_at
"2026-02-16T13:41:28.737439+00:00"
2/16/2026, 1:41:28 PM
model
"openai/gpt-5-mini-2025-08-07"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2067
moderation_latency
(null)
generation_time
35069
tokens_prompt
77
tokens_completion
1249
native_tokens_prompt
66
native_tokens_completion
3855
native_tokens_completion_images
0
native_tokens_reasoning
3456
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
0.0077265
router
(null)
provider_responses
0
id
"resp_01a18486129984660169931e64ded08190aad3fb008b7f0365"
status
200
is_byok
false
latency
115
endpoint_id
"c4f66d01-20b0-4c27-a225-438ea22fda43"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5-mini-2025-08-07"
api_type
"completions"
id
"gen-1771249252-hjO3YAEvG7CKLvDrLi0c"
upstream_id
"resp_01a18486129984660169931e64ded08190aad3fb008b7f0365"
total_cost
0.0077265
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
97.6276%
Matches word count
n/a
neededClean
false
words
207
0.0000%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 90.00%
neededClean
false
wordsTotal
207
wordsDialogue
0
48.8138%