NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-06-01KK1HT1PMMHGKA0WW5MQ5T9E5
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
11.8s
Cost:
$0.005025
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
97.63%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
67.2078%
Overall
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
“Bill, you cannot wear the clown shoes. They squeak guilt.” “They squeak whimsy, Molly. Children trust whimsy.” “Children also trust balloons, and I’m not robbing a piñata in size fourteen slippers.” “You’ve got no vision. We stroll in as entertainers, distract the parents, sweep the presents, vanish before the cake notices.” “The cake notices everything. Fondant is basically surveillance.” “Fine. Magician and assistant.” “I’m not sawing anyone in half before noon.” “Then caterers.” “In these coats? We look like depressed ushers.” “You always crush possibility.” “I preserve freedom, Bill. There’s a difference.” “Right. New plan. We arrive late, carrying a giant gift.” “Empty?” “Obviously empty. We swap ours for all of theirs.” “That’s not a plan. That’s seasonal etiquette with felonies.” “Thank you.” “We still need cover.” “Grandparents.” “You’re thirty-seven.” “In low light, I’m seventy.” “No, in low light you’re damp.” “Harsh. All right, face painter?” “You can’t paint.” “I can do tiger stripes and emotional damage.” “Listen. We go in as the puppet show. Kids stare, parents film, I bag the presents, you keep talking.” “My finest crime skill.” “You were born monologuing.” “What’s our exit?” “Through the bouncy castle.” “That’s genius.” “It’s soft genius.” “And if we’re caught?” “We say we’re teaching sharing.” “Villainous.” “Educational.”
created_at
"2026-03-06T12:26:25.372Z"
3/6/2026, 12:26:25 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
395
moderation_latency
(null)
generation_time
11784
tokens_prompt
77
tokens_completion
465
native_tokens_prompt
66
native_tokens_completion
324
native_tokens_completion_images
0
native_tokens_reasoning
0
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
0.005025
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_032f5e7ecba7ce880169aac7f172108193b69424a588034254"
is_byok
false
latency
164
model_permaslug
"openai/gpt-5.4-20260305"
provider_name
"OpenAI"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1772799985-7hZsQa0EwfK1Em9QoPzO"
upstream_id
"resp_032f5e7ecba7ce880169aac7f172108193b69424a588034254"
total_cost
0.005025
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
97.63%
Matches word count
n/a
neededClean
false
words
207
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
208
wordsDialogue
208
67.2078%