NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM0ACYZHJ7J2XPZTYG6QDTBF
Run Details
Model:
GPT-5.4 Mini
Test:
Dialogue tags
Scenario:
Write unattributed dialogue
Duration:
2.3s
Cost:
$0.001495
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 cheerful jingle as Clara stepped inside, bringing a gust of October air with her. Warmth wrapped around her at once, thick with the smell of cinnamon, butter, and fresh bread. “You’re late.” “I know.” “Three minutes.” “You counted?” “I also counted the croissants. One batch short.” Clara set her bag on the counter and peered toward the glass case. “That’s impossible. I baked enough for the morning rush.” “Then explain why Mrs. Pritchard just left with her usual six and Mr. Bellamy took the last cheese scone.” “He took your cheese scone?” “He called it a public service.” A laugh slipped out before Clara could stop it. Behind the counter, the flour-dusted register sat open, and the tin of ginger snaps was empty except for crumbs. “Did anyone else come in?” “Only the mayor, looking suspiciously pleased with himself.” “That’s never a good sign.” “No. Especially since he asked whether you were ‘the sort who keeps secrets.’” Clara froze. “What did you say?” “That secrets were best served with coffee, and he should order a cinnamon roll if he wanted any from me.” Outside, the rain tapped softly against the window. Clara glanced at the back door, where a muddy footprint marked the tile. “Well,”
she said
, lowering her voice, “I think somebody’s been in here before I arrived.” “Good.” “Good?” “Now we have an excuse to open the almond tarts early.”
created_at
"2026-03-18T11:13:23.967Z"
3/18/2026, 11:13:23 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
378
moderation_latency
(null)
generation_time
2129
tokens_prompt
81
tokens_completion
421
native_tokens_prompt
67
native_tokens_completion
321
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
0.00149475
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_0a2e94df02fb55ca0169ba88d40a1081978e818b5d825c95fb"
is_byok
false
latency
204
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-1773832403-ORGazVnxXtvte4d2pIaU"
upstream_id
"resp_0a2e94df02fb55ca0169ba88d40a1081978e818b5d825c95fb"
total_cost
0.00149475
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
"she said"