NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHKANVKQBC6CJDXWJQSZB5T2
Run Details
Model:
GPT-5 Mini
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
1m 23s
Cost:
$0.016969
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: "Kids' party, cake fog, one way in. You distract with balloon trick, I snag the gifts." Bill: "Balloon trick? Last time you popped a poodle. Not touching pets." Molly: "This is latex-free. I'm bringing a... confetti generator." Bill: "Confetti generator? You'll blow our cover." Molly: "Only if you panic. You memorize escape route: north door, through the bouncy castle." Bill: "Through the bouncy castle? I get stuck in sleeves, remember. Sleeveless." Molly: "We tie your sleeves. You wear the clown backpack disguise." Bill: "Clown backpack? I hate sequins." Molly: "Sequins hide shine. You carry present-sack labeled 'Leftover PiƱata Parts.'" Bill: "Genius. Position me by the snack table; sugar attracts parents like wolves." Molly: "I'll be on cake patrol. When candles go 'whoosh,' we swoop." Bill: "What about the baby? The baby always cries and ruins timing." Molly: "We'll recruit the birthday kid with a promise: extra balloon for a secret handshake." Bill: "Fine. But no glitter. That one was a war." Molly: "Deal. Tonight, we rob Santa-lite and call it charity." Bill: "What about a lookout? Ms. Thompson snoops." Molly: "Lookout will be the gardener. He hates rakes." Bill: "He owes you one? Bribed with leftover cupcakes?" Molly: "Perfect."
created_at
"2026-02-16T13:38:07.899749+00:00"
2/16/2026, 1:38:07 PM
model
"openai/gpt-5-mini-2025-08-07"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
31371
moderation_latency
(null)
generation_time
82935
tokens_prompt
77
tokens_completion
1091
native_tokens_prompt
66
native_tokens_completion
8476
native_tokens_completion_images
0
native_tokens_reasoning
8128
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
0.0169685
router
(null)
provider_responses
0
id
"resp_079b3452de694b800169931d6c3cd08193abbca6e385bcd7f4"
status
200
is_byok
false
latency
147
endpoint_id
"c4f66d01-20b0-4c27-a225-438ea22fda43"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5-mini-2025-08-07"
api_type
"completions"
id
"gen-1771249004-VJeZZ9w3ex9SRyN6tcwH"
upstream_id
"resp_079b3452de694b800169931d6c3cd08193abbca6e385bcd7f4"
total_cost
0.0169685
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
99.9983%
Dialogue to Total Word Ratio
Ratio: 90.64%, Deviation: 0.64%
neededClean
false
wordsTotal
203
wordsDialogue
184
99.9587%