NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM07MZA3TDE2QD4SWQS3FE31
Run Details
Model:
GPT-5.4 Mini (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
8.8s
Cost:
$0.009378
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
96.14%
Dialogue to Total Word Ratio
(details)
98.0680%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 50% 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 unfolded the party invitation like a treasure map. Across the street, a bouncy castle shuddered in the dusk while tinny music leaked from the community hall. Bill peered through binoculars he insisted made him look professional. “We are definitely too old for face paint,” he said. “Speak for yourself,” Molly said, adjusting her fake mustache. “I’m already emotionally committed to the pirate theme.” Bill frowned. “We’re stealing presents from seven-year-olds. Shouldn’t we feel at least a little villainous?” “Bill, we’re thieves, not philosophers.” He checked his clipboard. “Step one: enter as clowns.” “I hate clowns.” “You said that last time. Last time went badly.” “Last time the goose attacked first.” From inside, children shrieked with delight, and the hall doors swung open for a magician carrying a suspiciously large rabbit. Molly smiled. “Perfect distraction.” Bill grinned despite himself. “You mean our grand criminal opportunity?” “No,” said Molly, tightening her gloves. “I mean the cake table.” Bill gasped. “Molly, that is diabolical.” She winked. “We’ll take the presents on the way out.” “Remember,” Molly whispered, “no harming the children. We’re professionals.” Bill nodded. “Of course. We only target gifts, not innocence.” “And wrapping paper,” she said. “It’s always ours, darling.”
created_at
"2026-03-18T10:25:20.718Z"
3/18/2026, 10:25:20 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
8255
moderation_latency
(null)
generation_time
8639
tokens_prompt
77
tokens_completion
496
native_tokens_prompt
66
native_tokens_completion
2073
native_tokens_completion_images
0
native_tokens_reasoning
1766
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
0.009378
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_064531b816ab491f0169ba7d90c6c88190908c99e6030f6afa"
is_byok
false
latency
189
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-1773829520-2NsH5iyy7O5z5ALxhKfU"
upstream_id
"resp_064531b816ab491f0169ba7d90c6c88190908c99e6030f6afa"
total_cost
0.009378
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
200
96.14%
Dialogue to Total Word Ratio
Ratio: 54.46%, Deviation: 4.46%
neededClean
false
wordsTotal
202
wordsDialogue
110
98.0680%