NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHKAYH4PQ7JJS7N7GF1CQV1S
Run Details
Model:
GPT-5
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
1m 10s
Cost:
$0.034272
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: "Step one, we blend in like fondant ninjas; that means smiles, napkins, and zero sermons on dental hygiene." Bill: "Important question: what's our cover? I can only juggle cookies if I’m eating them simultaneously." Molly: "We’re distant cousins of the cake, here to offer moral support until it’s sliced." Bill: "So we orbit the frosting, nod at balloons, and quietly harvest gifts like benevolent raccoons?" Molly: "Precisely, only less crinkly. Remember, the squeak of wrapping paper is basically an alarm." Bill: "Then we speak in whispers so soft even sprinkles can’t hear us." Molly: "Also, keep eye contact with the nearest toddler; they can smell uncertainty." Bill: "I’ll project confident nap energy. Kids respect a man who believes in naps." Molly: "If cornered, compliment the piñata’s leadership qualities and pivot." Bill: "Pivot to where?" Molly: "To the refreshment table, the Switzerland of parties. Neutral, sticky, and diplomatically useful." Bill: "And the exit?" Molly: "Exists in our hearts until dessert. Never antagonize a child mid-cupcake." Bill: "Understood. If asked to sing, I only know the chorus of Happy Birthday seventeen times." Molly: "Perfect. Monotony is camouflage." Bill: "What if a parent suspects?" Molly: "We return decoys and glow with altruism."
created_at
"2026-02-16T13:42:38.174922+00:00"
2/16/2026, 1:42:38 PM
model
"openai/gpt-5-2025-08-07"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
13768
moderation_latency
(null)
generation_time
68768
tokens_prompt
77
tokens_completion
1151
native_tokens_prompt
66
native_tokens_completion
3419
native_tokens_completion_images
0
native_tokens_reasoning
3072
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
0.0342725
router
(null)
provider_responses
0
id
"resp_0f8b4d5ed0b07e0f0169931e886f4c8193aa0adb89810f613c"
status
200
is_byok
false
latency
168
endpoint_id
"7c2f859a-7890-4e8e-b1de-1cd1c0a800b4"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5-2025-08-07"
api_type
"completions"
id
"gen-1771249288-zuOpB0knq9BHeXXH024M"
upstream_id
"resp_0f8b4d5ed0b07e0f0169931e886f4c8193aa0adb89810f613c"
total_cost
0.0342725
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
99.9990%
Matches word count
n/a
neededClean
false
words
201
99.9434%
Dialogue to Total Word Ratio
Ratio: 91.54%, Deviation: 1.54%
neededClean
false
wordsTotal
201
wordsDialogue
184
99.9712%