NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM09YJQ0TR68TQF6KZ0CZ77P
Run Details
Model:
GPT-5.4 Mini
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
2.6s
Cost:
$0.001525
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
20.19%
Matches word count
(details)
95.64%
Dialogue to Total Word Ratio
(details)
57.9155%
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 you remembered the map,” Molly whispered. “I remembered the map,” Bill whispered back. “I also remembered the balloons are terrifying.” “They’re not terrifying.” “They’re aggressive decorations.” “Focus. We go in through the side gate.” “The side gate has a sign that says ‘Do Not Enter.’” “Exactly. No one ever guards the obvious thing.” “Sounds like the sort of logic that ends with us in a hedge.” “Makes it exciting.” “Does it?” “Makes it profitable. Can you lift the lock?” “I can lift the lock, yes. I can also trip over a crumb.” “That’s why I’m the brains.” “That’s not reassuring.” “Inside, we blend in.” “With what?” “With the other adults.” “I don’t own a paper cone hat.” “Borrow one.” “From who?” “From destiny.” “Right. Destiny. The six-year-old in the crown?” “Don’t point. Act innocent.” “I’m wearing black gloves.” “So? Children love mystery.” “Children love cake.” “Perfect. We’ll let them focus on cake while we take the presents.” “All of them?” “All of them.” “Even the giant one with the wheels?” “Especially the giant one with the wheels.” “That one looks expensive.” “That one looks portable.” “Fine. We grab, we run, we vanish.” “Like professionals.” “Like professionals,” Bill said. “But if anyone asks, I was never here.” “Molly snorted. “Bill, you’re wearing clown socks.” “Exactly,” he said. “Invisible.”
created_at
"2026-03-18T11:05:32.654Z"
3/18/2026, 11:05:32 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
825
moderation_latency
(null)
generation_time
2481
tokens_prompt
77
tokens_completion
473
native_tokens_prompt
66
native_tokens_completion
328
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
0.0015255
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_0fe46d021514baa70169ba86fcb5d88193b725a054ae8eb7d5"
is_byok
false
latency
198
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-1773831932-rVQdw62spgRI4oUyRq7R"
upstream_id
"resp_0fe46d021514baa70169ba86fcb5d88193b725a054ae8eb7d5"
total_cost
0.0015255
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
20.19%
Matches word count
n/a
neededClean
false
words
220
95.64%
Dialogue to Total Word Ratio
Ratio: 94.59%, Deviation: 4.59%
neededClean
false
wordsTotal
222
wordsDialogue
210
57.9155%