NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHK9RN09G2NZXBNN0ZKQ3Z08
Run Details
Model:
GPT-5 Nano
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
1m 39s
Cost:
$0.003782
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
Moonlight drips over the cul-de-sac where party favors glimmer like stars. Molly and Bill hunch behind a hedge, noses wrinkled at the scent of sugar paint and hot cocoa. A sprinkler squeaks, and laughter climbs the fence. Molly digs a finger through her hoodie pocket, pulling out a crumpled blueprint that looks like a doodle. Bill checks an invisible watch, muttering about timing, cake layers, and the weight of a dozen presents. Their plan unfolds in whispers, ridiculous yet precise, comical pact between two professionals pretending to be amateurs tonight. Molly: "We case the party, Bill, before the balloons explode tonight." Bill: "I remember snacks attract guards; we can glide past them." Molly: "Ignore guards, borrow a flashlight, and vanish with glitter tonight." Bill: "But door hinges whisper; we need a unicorn key, obviously." Molly: "Children's party is a fortress; we enter through cake doors." Bill: "The presents sing when shaken; silence is our best disguise." Molly: "Plan B involves glitter, confetti, and a really tiny heist." Bill: "We rehearse in the delivery van like grown-up toddlers tonight." Molly: "If we trigger zero alarms, we split with the glitter." Bill: "Deal; then we bake ourselves into the exit like elves."
created_at
"2026-02-16T13:22:26.268126+00:00"
2/16/2026, 1:22:26 PM
model
"openai/gpt-5-nano-2025-08-07"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2234
moderation_latency
(null)
generation_time
98382
tokens_prompt
77
tokens_completion
1773
native_tokens_prompt
66
native_tokens_completion
9447
native_tokens_completion_images
0
native_tokens_reasoning
9152
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
0.0037821
router
(null)
provider_responses
0
id
"resp_056ebd99b2e3c3f901699319af44f481978ac0bfd33c298214"
status
200
is_byok
false
latency
121
endpoint_id
"50329d77-04e1-4979-a184-c33030289476"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5-nano-2025-08-07"
api_type
"completions"
id
"gen-1771248047-Id6TtdV29c6byNC6vEig"
upstream_id
"resp_056ebd99b2e3c3f901699319af44f481978ac0bfd33c298214"
total_cost
0.0037821
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
100.0000%
Dialogue to Total Word Ratio
Ratio: 49.75%, Deviation: 0.25%
neededClean
false
wordsTotal
203
wordsDialogue
101
99.9595%