NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM099Y63CWV5YGEB391EREQJ
Run Details
Model:
GPT-5.4 Mini
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
2.0s
Cost:
$0.001413
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
68.10%
Matches word count
(details)
0.08%
Dialogue to Total Word Ratio
(details)
34.0902%
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 peered through the bakery window at the rainbow balloons wobbling over the hall next door. “There,” she whispered, “a fortress of frosting and weak security.” Bill adjusted his cap. “You keep calling every place a fortress.” “Because every place has a door,” Molly said, “and every door has a fool nearby.” Bill squinted at the invitation he’d stolen from the trash. “It says the party starts at two, includes games, cake, and a magician.” “Perfect,” Molly said. “Magicians love distraction. Children love noise. Parents love sitting down. We arrive as harmless volunteers.” Bill frowned. “I don’t know how to be harmless.” “You already are,” Molly said. “Just hold a tray.” He nodded. “And then?” “We blend in, praise the balloon arch, compliment the puppet theater, and wait for the present pile to grow complacent.” Bill looked nervous. “What if a kid asks questions?” Molly smiled. “Then we say we are the Official Gift Relocation Team.” “That sounds fake.” “Exactly,” Molly said. “Fake is the foundation of confidence.” A burst of children’s laughter shook the hall. Bill swallowed. “How do we carry them out?” Molly tapped the shopping cart hidden behind the hedge. “In batches.” Bill blinked. “That’s your plan?” Molly grinned. “Bill, darling, every great theft begins with cake and ends with logistics.”
created_at
"2026-03-18T10:54:16.272Z"
3/18/2026, 10:54:16 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
409
moderation_latency
(null)
generation_time
1873
tokens_prompt
77
tokens_completion
395
native_tokens_prompt
66
native_tokens_completion
303
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
0.001413
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_0c4e3b313ef919340169ba8458556481949356830a9d1bc0ad"
is_byok
false
latency
197
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-1773831256-rCswZkiwfMkdOKyGOqFx"
upstream_id
"resp_0c4e3b313ef919340169ba8458556481949356830a9d1bc0ad"
total_cost
0.001413
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
68.10%
Matches word count
n/a
neededClean
false
words
214
0.08%
Dialogue to Total Word Ratio
Ratio: 66.36%, Deviation: 16.36%
neededClean
false
wordsTotal
214
wordsDialogue
142
34.0902%