NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM0886CZTKGHWPJP9ZWGFKFQ
Run Details
Model:
GPT-5.4 Nano
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
1.8s
Cost:
$0.000416
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.18%
Dialogue to Total Word Ratio
(details)
0.0905%
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 crouched behind a potted fern, eyeing the glittering front door like it was a vault with frosting instead of steel. Bill nudged a toolbox with his boot, then froze as a wind chime clinked somewhere inside. “Okay,” he whispered, “tell me you hear that too.” Molly lifted a finger to her lips, then pointed at the doormat. The welcome mat read YOU ARE ENTERING A FUN ZONE in cheerful crayon, which felt like a threat disguised as a compliment. Bill leaned in, pretending not to tremble. “That sign is basically a security system,” he said. Molly snorted softly. “It’s security theater.” She slid a tiny lockpick set from her sleeve, careful as a surgeon, except her hands looked like they’d stolen their confidence from a circus. Bill examined the handle, noting a sticker shaped like a smiling dinosaur. “If that thing starts roaring, we’re doomed,” he murmured. Molly grinned. “Then we’ll distract it with our charm.” She tapped her earpiece; static answered like distant laughter. “We’re here for the presents,” Bill reminded her, as if the entire house hadn’t already screamed it in wrapping paper. Molly nodded toward the living room windows, where twinkling lights bobbed amid cardboard and bows. “Plan is simple,” she said. “We go in, we take everything, we leave a single toy behind.” Bill blinked. “Why?” Molly’s eyes glittered. “To keep the kids from calling the cops.”
created_at
"2026-03-18T10:35:50.573Z"
3/18/2026, 10:35:50 AM
model
"openai/gpt-5.4-nano-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
437
moderation_latency
(null)
generation_time
1625
tokens_prompt
77
tokens_completion
368
native_tokens_prompt
66
native_tokens_completion
322
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
0.0004157
router
(null)
provider_responses
0
endpoint_id
"0c835f2e-c18d-4e8c-b245-e1e3bd08b97f"
id
"resp_048ec593969737dd0169ba8006a0d0819683f4ec62763ab282"
is_byok
false
latency
177
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-1773830150-TUmlgEB1u6LyQ5WvOqCq"
upstream_id
"resp_048ec593969737dd0169ba8006a0d0819683f4ec62763ab282"
total_cost
0.0004157
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
232
0.18%
Dialogue to Total Word Ratio
Ratio: 25.86%, Deviation: 15.86%
neededClean
false
wordsTotal
232
wordsDialogue
60
0.0905%