NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM08K0J860KV5GJR38RJZD0M
Run Details
Model:
GPT-5.4 Mini (Reasoning)
Test:
Dialogue tags
Scenario:
Write unattributed dialogue
Duration:
3.8s
Cost:
$0.002215
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
60.65%
Count dialogue tags
(details)
Model input
User/Human
You are an expert novel writer in the cozy mystery genre. Write a scene of about 200 words, with dialogue between two characters. The dialogue should be unattributed, meaning the speaker is not explicitly mentioned. The scene should be set in a small town bakery. Only return the story, nothing else.
Model response
Messages
Meta
AI
The bell over the bakery door gave a bright little jingle as Nora came in from the misty street, bringing a gust of cold air and the smell of wet leaves. “You’re late,” Mrs. Finch called from behind the counter, where a row of pear tarts cooled beneath a wire rack. “I found a clue.” Mrs. Finch’s hands stilled over the pastry brush. “That sounds much less pleasant than the last time you said it.” “It was in the flour bin at the library fundraiser. A silver spoon. Not a regular one—engraved with a sunflower.” Mrs. Finch’s eyes narrowed. “Sunflowers don’t belong in flour bins.” “No. But they do belong on the old Hazelton tea service.” The baker glanced toward the back room, then lowered her voice. “You’ve been asking about that missing set all week.” “And you’ve been pretending you don’t care.” A thin line formed between Mrs. Finch’s brows, but the corners of her mouth twitched. “I care very much. Especially if someone has been sneaking into my bakery after closing.” Nora leaned closer. “You think it’s connected?” “I think,” Mrs.
Finch said
, lifting a tart and setting it on a plate, “that if someone hid a stolen spoon in my flour, they were either very foolish or very hungry.” Outside, the church clock struck noon. “Which do you prefer?” Mrs. Finch slid the tart across the counter. “In this town? Both usually lead to trouble. Eat first. Investigate after.”
created_at
"2026-03-18T10:41:45.046Z"
3/18/2026, 10:41:45 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3308
moderation_latency
(null)
generation_time
3559
tokens_prompt
81
tokens_completion
522
native_tokens_prompt
67
native_tokens_completion
481
native_tokens_completion_images
0
native_tokens_reasoning
159
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
0.00221475
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_0fd803f1f95f4e0b0169ba81691db08194826929d4dbcf809e"
is_byok
false
latency
174
model_permaslug
"openai/gpt-5.4-mini-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-1773830505-XTSGKtZ1bAEP8OfYm7VT"
upstream_id
"resp_0fd803f1f95f4e0b0169ba81691db08194826929d4dbcf809e"
total_cost
0.00221475
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
60.65%
Count dialogue tags
n/a
neededClean
false
dialogueTags
0
"Finch said"