NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-09-01KK9PTSXK5TQPRV2BVAYMY324
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write unattributed dialogue
Duration:
13.7s
Cost:
$0.006692
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
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 its usual cheerful jingle, though nothing felt cheerful beneath the warm haze of cinnamon and yeast. Morning light pooled across the glass cases, catching on sugared knots and apricot tarts, while behind the counter a tray of scones cooled beside the till. “You’re early.” “You’re open.” “Barely. The coffee’s still deciding whether it wants to be coffee.” “I’m not here for coffee.” “No one ever says that unless they’re here for trouble.” A pause settled between the wicker bread baskets. “Did you hear?” “In a town this size, I heard three versions before dawn. Which one do you mean?” “The one with Mr. Pembroke.” “The one where he slipped on the church steps, the one where he ran off with the librarian, or the one where he was found face-down in the alley behind the flower shop?” “The alley.” Flour dust drifted in the sunbeams like pale smoke. “They’re calling it a fall.” “And you don’t believe that.” “He came in yesterday for a blackberry turnover. Paid with exact change. Hands steady. Shoes polished. A man planning to die in an alley doesn’t ask whether the blackberries were local.” “Maybe he changed his mind after the turnover.” “No. But he did leave this.” A folded paper appeared on the counter, tucked beneath the edge of a cherry pie stand. “That’s a receipt.” “Turn it over.” “…Maple Lane. Half past nine.” “Exactly.” “Why give it to you?” “He didn’t. He hid it in the tip jar.”
created_at
"2026-03-09T16:28:08.51Z"
3/9/2026, 4:28:08 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4828
moderation_latency
(null)
generation_time
13589
tokens_prompt
81
tokens_completion
564
native_tokens_prompt
67
native_tokens_completion
435
native_tokens_completion_images
0
native_tokens_reasoning
103
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
0.0066925
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_07b84d2f7dcd313e0169aef518a21881979b3ababf00080c9d"
is_byok
false
latency
202
model_permaslug
"openai/gpt-5.4-20260305"
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-1773073688-DrMcIGthXGKJ4gjqNLwQ"
upstream_id
"resp_07b84d2f7dcd313e0169aef518a21881979b3ababf00080c9d"
total_cost
0.0066925
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Count dialogue tags
n/a
neededClean
false
dialogueTags
(empty)