NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM09EYT8SYRT1DASMYEPE131
Run Details
Model:
GPT-5.4 Nano
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
2.5s
Cost:
$0.000472
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.02%
Dialogue to Total Word Ratio
(details)
0.0102%
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 checked the cuff of her black gloves and whispered, “Okay, Bill, if we get caught, I’m blaming your haircut.” Bill crouched behind a shrub, peering at the glowing windows. “My haircut is a professional choice, thank you. Besides, children are easy targets.” Molly snorted. “Easy targets? They’re basically tiny security contractors. They squeal, they report, they judge.” Bill tapped his earpiece. “We don’t steal from kids, we steal from joy. Big difference.” Molly slid a small magnet along the porch rail. “Tell that to the police when they ask why you’re swiping Paw Patrol.” Bill leaned in. “Listen, the presents are all stacked by the tree. We take the shiny ones, leave the noisy ones. Mercy.” Molly grinned. “Mercy? You’re a monster with morals.” Bill nodded solemnly. “I have a code. Step one: no stepping on Lego.” Molly’s eyes widened. “You stepped on Lego last time.” Bill sighed. “It was one Lego. Technically, it attacked me.” Molly opened the gate with a soft click. “Fine. Step two: quiet entry, dramatic exits.” Bill raised a finger. “Dramatic exits are non-negotiable.” Molly hissed, “The party starts in three minutes.” Bill whispered, “Then we move like ninjas.” Molly glanced at his sneakers. “Ninjas don’t squeak.” Bill shrugged. “Neither do I, unless the air is full of whines and allergies.” Molly rolled her eyes. “Focus. We’re taking presents, not starting a memoir.” Bill took a breath. “Together, we become legend.” Molly smiled. “Or a cautionary tale for parents.”
created_at
"2026-03-18T10:57:00.758Z"
3/18/2026, 10:57:00 AM
model
"openai/gpt-5.4-nano-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
443
moderation_latency
(null)
generation_time
1951
tokens_prompt
77
tokens_completion
410
native_tokens_prompt
66
native_tokens_completion
367
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
0.00047195
router
(null)
provider_responses
0
endpoint_id
"0c835f2e-c18d-4e8c-b245-e1e3bd08b97f"
id
"resp_00bb1a227f831c070169ba84fcd1248197b1d73e2e210466af"
is_byok
false
latency
182
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-1773831420-GgPT2RAh1Qkqgn0WRh0k"
upstream_id
"resp_00bb1a227f831c070169ba84fcd1248197b1d73e2e210466af"
total_cost
0.00047195
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
245
0.02%
Dialogue to Total Word Ratio
Ratio: 67.07%, Deviation: 17.07%
neededClean
false
wordsTotal
246
wordsDialogue
165
0.0102%