NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHK9VQGCGQ1A9880FZDE5G2X
Run Details
Model:
GPT-5
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
56.4s
Cost:
$0.039152
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: The mission is elementary school level, which is terrifyingly advanced. Bill: Children are ruthless negotiators. Bring stickers, surrender pride, acquire presents. Molly: Disguises? Bill: Clown and magician. Juggle alibis; I pull excuses from hats. Molly: I juggle bowling pins and panic. Fine. Entry point? Bill: Bounce house. We slide under Velcro democracy. Molly: What about the dragon-shaped cake sentinel? Bill: I respect icing dragons. Bribe it with forks and deniability. Molly: Distraction plan? Bill: Musical chairs with only one chair; chaos blossoms, parents philosophize. Molly: While they debate fairness, we relocate gift ecosystem into our ethical custody. Bill: Exactly. Gift bags are our tiny loot satchels; wrapping paper is camouflage. Molly: What if a toddler fingerprints us with frosting? Bill: We claim it's avant-garde makeup, influencer tier: Sprinklescore. Molly: Escape route? Bill: Piñata evacuation. We shout, 'Candy migration!' and swim downstream. Molly: Contingency? Bill: If cornered, we perform a heartfelt puppet show about sharing. Molly: Morally ambiguous sharing. Bill: Educational, even. We leave one present: a note saying thank you. Molly: Signed by Santa’s Logistics Division, Seasonal Redistribution Wing. Bill: Beautiful. Synchronize watches? Molly: Mine has unicorns instead of numbers. Bill: Then at high noon-ish sparkle, we heist politely.
created_at
"2026-02-16T13:23:24.664451+00:00"
2/16/2026, 1:23:24 PM
model
"openai/gpt-5-2025-08-07"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
7775
moderation_latency
(null)
generation_time
56254
tokens_prompt
77
tokens_completion
1007
native_tokens_prompt
66
native_tokens_completion
3907
native_tokens_completion_images
0
native_tokens_reasoning
3520
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
0.0391525
router
(null)
provider_responses
0
id
"resp_03d46e06714886ab0169931a141f188194ab3aac94d53dcad0"
status
200
is_byok
false
latency
164
endpoint_id
"7c2f859a-7890-4e8e-b1de-1cd1c0a800b4"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5-2025-08-07"
api_type
"completions"
id
"gen-1771248148-TjCYsvKSgYIrTYBsab2j"
upstream_id
"resp_03d46e06714886ab0169931a141f188194ab3aac94d53dcad0"
total_cost
0.0391525
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
99.9840%
Matches word count
n/a
neededClean
false
words
202
0.0000%
Dialogue to Total Word Ratio
Ratio: 2.48%, Deviation: 87.52%
neededClean
false
wordsTotal
202
wordsDialogue
5
49.9920%