NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM09RZSH25NN4YMV7BFD5N1N
Run Details
Model:
GPT-5.4 Nano (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
33.8s
Cost:
$0.006543
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
100.0000%
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 and Bill slid through the hedge gap like two cupcakes. Spotlights from the backyard bounced over a castle of gift boxes, each ribbon winking as if it had an opinion. The party-goers milled inside, chasing helium balloons and arguing about who got to be captain of the game. Bill clicked his silent watch; it blinked a skull, then went dark. “Shh, Bill, shh!” Molly mouthed without sound. “Presents first.” The plan was simple: steal everything before anyone noticed the panic behind the tissue paper. Bill leaned toward the front door, where a battery-operated clown horn sat on a pedestal labeled WELCOME, PLEASE LAUGH. “Keys, Bill?” he asked. Molly patted her ankle and held up a laminated card that read EMERGENCY SOCK KEY. “Sock key.” She winked, as though footwear could be trusted with felony paperwork. Bill raised a finger at the motion sensor. “No alarms, please.” The horn’s speaker crackled, practicing giggles for them. Molly adjusted her clown mask, breathing in frosting-scented air. “Impersonate clowns.” Bill hauled a wagon toward the living-room, wheels squeaking like mice. “Grab them all.” Molly paused, counting boxes. “Run at midnight.” Outside, the pinata gong sounded, and their shadows scuttled toward the gift pile.
created_at
"2026-03-18T11:02:29.436Z"
3/18/2026, 11:02:29 AM
model
"openai/gpt-5.4-nano-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
33198
moderation_latency
(null)
generation_time
33606
tokens_prompt
77
tokens_completion
1416
native_tokens_prompt
66
native_tokens_completion
5224
native_tokens_completion_images
0
native_tokens_reasoning
4934
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
0.0065432
router
(null)
provider_responses
0
endpoint_id
"0c835f2e-c18d-4e8c-b245-e1e3bd08b97f"
id
"resp_052618fc3455777e0169ba864581dc819083432e2a01066771"
is_byok
false
latency
220
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-1773831749-vBcCR2cE7GhQQhD4mtGm"
upstream_id
"resp_052618fc3455777e0169ba864581dc819083432e2a01066771"
total_cost
0.0065432
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
200
100.00%
Dialogue to Total Word Ratio
Ratio: 9.80%, Deviation: 0.20%
neededClean
false
wordsTotal
204
wordsDialogue
20
100.0000%