NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM099D4TWKG1NJ7M5K1H0WWK
Run Details
Model:
GPT-5.4 Nano (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
28.6s
Cost:
$0.007637
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 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
Outside the house, Molly used the hedge as a runway, leaving mint crumbs. Bill followed, dragging a toolbox that rattled like coins in a sock. Inside the living room, lights flashed over frosting and a mountain of presents. Molly checked her stolen badge: BIRTHDAY POLICE, glitter font. Bill rolled his shoulders and aimed a glow-stick at the side door sensor. It blinked, then went green, as if the house approved. They slid through the opening and crouched by the gift table. The bows looked hungry, and Bill’s hands twitched toward crime. A timer in the hall hummed, counting down panic. "Bill, we only take presents, not toys already hugged." "Relax, I’ll lift gently; I’m basically a librarian for thieves." "If a kid asks questions, you answer with more questions." "I can do that; I’m great at interrogating balloons." "Bag duty for you, hat duty for me, okay?" "My hat is ready; it screams 'totally innocent' loudly." "Which presents first, the big dinosaur or the tiny trains?" "Start with dinosaurs; they distract adults with their roar." "When the timer hits seven, we sprint like birthday confetti." "And if we’re caught, we blame the cake committee!" "Now pick, before squeaky-dog patrols again."
created_at
"2026-03-18T10:53:58.823Z"
3/18/2026, 10:53:58 AM
model
"openai/gpt-5.4-nano-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
28130
moderation_latency
(null)
generation_time
28437
tokens_prompt
77
tokens_completion
1132
native_tokens_prompt
66
native_tokens_completion
6099
native_tokens_completion_images
0
native_tokens_reasoning
5809
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
0.00763695
router
(null)
provider_responses
0
endpoint_id
"0c835f2e-c18d-4e8c-b245-e1e3bd08b97f"
id
"resp_0bf20d8358480dba0169ba8446dfc88190b25d706a55a597bd"
is_byok
false
latency
190
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-1773831238-CpIwERoFCfY2Bl3hzZ4S"
upstream_id
"resp_0bf20d8358480dba0169ba8446dfc88190b25d706a55a597bd"
total_cost
0.00763695
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: 50.00%, Deviation: 0.00%
neededClean
false
wordsTotal
202
wordsDialogue
101
100.0000%