NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM0A285QYW4HJAKHAKX45A7Z
Run Details
Model:
GPT-5.4 Nano
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
2.0s
Cost:
$0.000436
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
14.30%
Matches word count
(details)
73.29%
Dialogue to Total Word Ratio
(details)
43.7967%
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
The alley behind the pastel house smelled like frosting and bad decisions. Molly crouched by the hedge, whispering, “Bill, if a toddler sees us, we’re done.” Bill adjusted his grappling hook like it was a résumé. “Relax, I’ve practiced being invisible.” “You practiced being loud,” Molly said. “I practiced being confident.” “Confidence is how you get caught.” Bill held up a laminated map. “I mapped the security cameras.” “Those are toy dinosaurs,” Molly replied. “Technically, they still count as cameras.” Molly snorted, then slid a tiny crowbar from her pocket. “What’s our entry plan?” Bill grinned. “We wait for the piñata.” “That’s not a plan, that’s chaos.” “Chaos is my specialty.” Molly leaned closer. “Remember: we only steal the presents.” Bill clasped his hands. “Right. All presents. No stealing the juice boxes.” Molly stared. “Why would you steal juice boxes?” “Because they’re everywhere, and they whisper, ‘Bill, you deserve happiness.’” Molly glanced at the window. Kids were giggling. “If you hear your name, that’s not the juice boxes.” Bill’s eyes widened. “It might be the juice boxes.” Molly hissed, “Bill.” Bill sighed. “Fine. No juice boxes. Just the presents.” Molly nodded once. “Then on three, we go.” Bill whispered, “One.” Molly whispered, “Two.” Bill whispered, “I’m already regretting this.” “Three,” Molly said, and the hedge swallowed them like a guilty secret.
created_at
"2026-03-18T11:07:32.93Z"
3/18/2026, 11:07:32 AM
model
"openai/gpt-5.4-nano-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
565
moderation_latency
(null)
generation_time
1863
tokens_prompt
77
tokens_completion
398
native_tokens_prompt
66
native_tokens_completion
338
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
0.0004357
router
(null)
provider_responses
0
endpoint_id
"0c835f2e-c18d-4e8c-b245-e1e3bd08b97f"
id
"resp_0677de8decc692060169ba877500cc8190bea7c1a314274c34"
is_byok
false
latency
304
model_permaslug
"openai/gpt-5.4-nano-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-1773832052-UKzyfCzqX608KnCRqce6"
upstream_id
"resp_0677de8decc692060169ba877500cc8190bea7c1a314274c34"
total_cost
0.0004357
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
14.30%
Matches word count
n/a
neededClean
false
words
221
73.29%
Dialogue to Total Word Ratio
Ratio: 57.47%, Deviation: 7.47%
neededClean
false
wordsTotal
221
wordsDialogue
127
43.7967%