NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM09R3STXRNK4G9Z6F71Z0NT
Run Details
Model:
GPT-5.4 Mini (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
3.0s
Cost:
$0.001566
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
99.92%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
68.3535%
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, tell me again why we’re robbing a children’s party.” “Because, Molly, adults lock their wallets, children leave loot under wrapping paper.” “You call a bicycle and three dinosaurs ‘loot’?” “In our line of work, yes.” “I thought our line of work was tasteful burglary, not festive vandalism.” “Festive vandalism is just burglary with streamers.” “Mmm. Any security?” “Two parents, one clown, and a bouncy castle.” “That’s not security.” “It is if the clown is aggressive.” “Makes sense. What’s the plan?” “We enter as entertainers.” “I’m not entertaining anyone.” “You can juggle.” “I can juggle apples.” “Children love apples.” “Children also bite.” “Then keep your distance. I’ll distract the adults.” “How?” “I’ll say we’re from the party-planning committee.” “Very convincing.” “I’ll wear a tie.” “You don’t own a tie.” “I’ll borrow one from the lost-and-found of society.” “Bill, are you sure we can carry all the presents?” “Of course.” “Through the hedge?” “Through the hedge.” “Won’t the wrapping paper rattle?” “Not if we stack the teddy bears on top.” “And if we get caught?” “Smile. No one suspects joy.” “You’re disgusting.” “Thank you. Ready?” “As ready as a thief can be.” “Then let’s go ruin childhood, elegantly.”
created_at
"2026-03-18T11:02:00.776Z"
3/18/2026, 11:02:00 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2635
moderation_latency
(null)
generation_time
2838
tokens_prompt
77
tokens_completion
553
native_tokens_prompt
66
native_tokens_completion
337
native_tokens_completion_images
0
native_tokens_reasoning
35
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
0.001566
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_06a23b7e66c12ade0169ba8628d4048197894d7eaa3a69a3c2"
is_byok
false
latency
170
model_permaslug
"openai/gpt-5.4-mini-20260317"
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-1773831720-nJbanEsBKEZW3MsZqPVV"
upstream_id
"resp_06a23b7e66c12ade0169ba8628d4048197894d7eaa3a69a3c2"
total_cost
0.001566
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
99.92%
Matches word count
n/a
neededClean
false
words
197
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
200
wordsDialogue
200
68.3535%