NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM09G4APNVKYZZPFBRZ7TX11
Run Details
Model:
GPT-5.4 Nano
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
2.8s
Cost:
$0.000451
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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 pressed her cheek to the window and peered inside the bright, sugar-scented chaos. Balloons bobbed like suspicious jellyfish. A glittery banner read HAPPY BIRTHDAY, as if the letters themselves were armed with frosting. Bill crouched beside her, juggling a roll of duct tape, a tiny wrench, and the confidence of a man who’d once tried to open a vending machine with a spoon. “Okay,” Bill whispered, “we take the presents, zip out, and nobody’s the wiser.” Molly traced a careful line on the fogged glass with a fingernail. “Nobody’s the wiser except the very observant three-year-olds.” Bill glanced at the yard where a clown-shaped pinwheel spun with terrifying enthusiasm. “Kids can’t be that observant.” Molly snorted. “They can spot a hiding spot from three rooms away. It’s basically their superpower.” Bill produced a laminated plan labeled Operation Blanket Fortress. “We’re using the shortcut through the playhouse. It’s made of plastic and courage.” Molly adjusted her mask, which was just a black beret pretending to be useful. “Fine. But if we bump the piñata, we’re paying for therapy.” Bill nodded, then paused as the front door clicked open somewhere beyond the wall. “You hear that?” Molly grinned, finally enjoying the mess. “That’s the sound of destiny—also, possibly, a parent looking for lost socks.” Bill whispered, “Can destiny at least leave us alone long enough to steal the presents?” “Only if the presents don’t squeak,” Molly said, and they slid into the night like two misbehaving shadows.
created_at
"2026-03-18T10:57:39.172Z"
3/18/2026, 10:57:39 AM
model
"openai/gpt-5.4-nano-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1089
moderation_latency
(null)
generation_time
2563
tokens_prompt
77
tokens_completion
424
native_tokens_prompt
66
native_tokens_completion
350
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
0.0004507
router
(null)
provider_responses
0
endpoint_id
"0c835f2e-c18d-4e8c-b245-e1e3bd08b97f"
id
"resp_0267f3abd85c20c50169ba85233c3081909445bdb52de84eb1"
is_byok
false
latency
181
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-1773831459-lmVUwwfckpb1hoLH73sP"
upstream_id
"resp_0267f3abd85c20c50169ba85233c3081909445bdb52de84eb1"
total_cost
0.0004507
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
247
0.00%
Dialogue to Total Word Ratio
Ratio: 39.44%, Deviation: 29.44%
neededClean
false
wordsTotal
251
wordsDialogue
99
0.0000%