NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM0ABXTD3DS2KMKB9S7JK9JX
Run Details
Model:
GPT-5.4 Nano
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
2.1s
Cost:
$0.000488
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
11.50%
Dialogue to Total Word Ratio
(details)
5.7475%
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 adjusted her beanie and peered at the balloon arch. “Okay, Bill, remember: no lasers, no grand monologues, and definitely no squeaky shoes.” Bill crouched by the hedge, holding a clipboard like it was a master key. “Squeaky shoes are my brand. Besides, the plan is foolproof: we enter through the side gate, blend in as ‘mysterious helpers,’ and—” Molly cut him off. “Helpers? Bill, we’re stealing presents from a children’s party. We’re not joining a daycare committee.” Bill grinned. “Morally ambiguous daycare committee.” They inched toward the door, where a sprinkler hissed rhythmically. “Do you hear that?” Molly whispered. “That’s the sound of my anxiety evaporating.” Bill held up a tiny smoke bomb labeled “CONFETTI.” “And this is the sound of your anxiety getting arrested.” Molly blinked. “Confetti bombs are for birthdays, not crimes.” “Technically,” Bill said, “I’m confusing the universe. Now—on three.” Molly nodded. “One, two—” A toddler’s laugh erupted from inside. “Abort?” Bill asked. “No,” Molly said, “we can’t abort. We’re basically ninjas with sticky fingers.” Bill checked his bag. “I’ve got gloves, duct tape, and a decoy toy duck.” Molly smirked. “A duck? That’s your intimidation?” Bill whispered fiercely, “Quack is terrifying.” The door creaked. Molly hissed, “Quiet, Bill.” Bill whispered back, “I’m so quiet I could be a tax audit.” Molly slipped inside. “Good. Now steal all the presents—politely.” Bill followed, whispering, “Nothing says ‘polite’ like silent thievery.”
created_at
"2026-03-18T11:12:50.01Z"
3/18/2026, 11:12:50 AM
model
"openai/gpt-5.4-nano-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
476
moderation_latency
(null)
generation_time
1933
tokens_prompt
77
tokens_completion
425
native_tokens_prompt
66
native_tokens_completion
380
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
0.0004882
router
(null)
provider_responses
0
endpoint_id
"0c835f2e-c18d-4e8c-b245-e1e3bd08b97f"
id
"resp_0d1c8910e4d6fdf60169ba88b214c08197a53b47021d853a40"
is_byok
false
latency
217
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-1773832370-OyW41V0CMQeBSeNzCy9K"
upstream_id
"resp_0d1c8910e4d6fdf60169ba88b214c08197a53b47021d853a40"
total_cost
0.0004882
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
235
11.50%
Dialogue to Total Word Ratio
Ratio: 62.13%, Deviation: 12.13%
neededClean
false
wordsTotal
235
wordsDialogue
146
5.7475%