NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM0ANBCNJPZ2EF1DK5A1YV12
Run Details
Model:
GPT-5.4 Mini
Test:
N-Length Sentences
Scenario:
Write sentences with 20 words each
Duration:
1.9s
Cost:
$0.000746
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
73.77%
Matches word count
(details)
Model input
User/Human
Please write the opening scene of a murder mystery where each sentence has exactly 20 words. Please make sure that each sentence has the same length. Only return the opening itself.
Model response
Messages
Meta
AI
Rain stitched the manor windows shut, while Inspector Vale climbed the marble steps, already sensing the house was holding breath. Inside, candles trembled over portraits, and Lady Merritt’s guests clustered silently, each avoiding the still figure beside the grand piano. The dead man wore a silver cufflink shaped like a raven, and his open eyes reflected a clock stopped at midnight. Vale knelt, noting the bloodless wound, the missing weapon, and the faint scent of lavender drifting from somewhere upstairs. No one spoke until the butler whispered that every door had been locked from within, moments before the scream. Vale looked around the room, studying trembling hands, composed faces, and one empty champagne glass with a lipstick mark.
created_at
"2026-03-18T11:17:58.82Z"
3/18/2026, 11:17:58 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
439
moderation_latency
(null)
generation_time
1758
tokens_prompt
46
tokens_completion
194
native_tokens_prompt
41
native_tokens_completion
159
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
0.00074625
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_066df81ba7e0610b0169ba89e6df4c819490ffa623782a8ca3"
is_byok
false
latency
178
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-1773832678-j1V7H7aIngLL14735j3N"
upstream_id
"resp_066df81ba7e0610b0169ba89e6df4c819490ffa623782a8ca3"
total_cost
0.00074625
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
73.77%
Matches word count
n/a
neededClean
false
sentences
6
wordCounts
0
20
1
20
2
21
3
19
4
19
5
19