NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM0A8DX7D2CREAR9NJ9RHS99
Run Details
Model:
GPT-5.4 Mini
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
2.6s
Cost:
$0.001669
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
93.65%
Matches word count
(details)
71.39%
Dialogue to Total Word Ratio
(details)
82.5210%
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
“Molly, you’re frowning.” “I’m thinking.” “That’s what worries me. Whenever you think, something expensive happens.” “Bill, hush. The party’s in the community hall.” “Children’s party,” Bill whispered, peering through the rain-streaked window. “Balloons, cake, paper hats, and probably tiny people with sticky fingers.” “Exactly.” “You say ‘exactly’ like that’s a plan.” “It is a plan. We go in, blend in, and leave with the presents.” Bill blinked. “At a children’s party?” “Everybody trusts a party clown.” “We are not clowns.” “We can be festive.” “We are thieves.” “Festive thieves.” Bill scratched his chin. “How many presents?” “All of them.” “All of them?” “All. Of. Them.” “Bit ambitious.” “Mildly ambitious.” “Criminally ambitious,” Bill said. “What’s the angle?” Molly lowered her voice. “I brought a stroller.” Bill stared. “Why have you brought a stroller?” “For cover. You push it, looking like a frazzled parent.” “I don’t look like a parent.” “Then look frazzled. Carry a balloon. Nobody suspects incompetence at a children’s party.” Bill sighed. “And the presents?” “In the hallway, unguarded.” “Ungaurded?” “Unsupervised.” “Same thing.” “Not to the law.” Bill nodded slowly. “And the children?” Molly smiled. “We distract them.” “With what?” She lifted a sack. “Party favors.” Bill squinted. “Those are just marbles.” “Exactly,” said Molly. “They’ll be thrilled.”
created_at
"2026-03-18T11:10:55.413Z"
3/18/2026, 11:10:55 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
766
moderation_latency
(null)
generation_time
2333
tokens_prompt
77
tokens_completion
481
native_tokens_prompt
66
native_tokens_completion
360
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
0.0016695
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_0afb52d038874cb20169ba883f7a7c8190a735164f35d569d1"
is_byok
false
latency
175
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-1773832255-SFvMIYl8yCAyBsvceqwr"
upstream_id
"resp_0afb52d038874cb20169ba883f7a7c8190a735164f35d569d1"
total_cost
0.0016695
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
93.65%
Matches word count
n/a
neededClean
false
words
209
71.39%
Dialogue to Total Word Ratio
Ratio: 82.38%, Deviation: 7.62%
neededClean
false
wordsTotal
210
wordsDialogue
173
82.5210%