NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-06-01KK1H0VFND0VRP8CQ4ZX0C2VW
Run Details
Model:
GPT-5.4
Test:
N-Length Sentences
Scenario:
Write sentences with 20 words each
Duration:
5.6s
Cost:
$0.001978
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
51.23%
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 needled the manor windows as guests arrived, clutching invitations embossed with black roses and the promise of reconciliation. Inside, candlelight trembled above silver trays while old grievances circulated politely, disguised as compliments, laughter, and inquiries about health. At precisely eight, the host descended smiling, though his valet noticed crimson on one cuff and fear behind charm. No one mentioned the storm-stranded bridge until thunder silenced conversation, and the library door eased open by itself. Then Mrs. Vale screamed, pointing inward, where Sir Alistair lay across the Persian rug, throat cut, wedding ring missing.
created_at
"2026-03-06T12:12:39.805Z"
3/6/2026, 12:12:39 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
438
moderation_latency
(null)
generation_time
5587
tokens_prompt
46
tokens_completion
165
native_tokens_prompt
41
native_tokens_completion
125
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
0.0019775
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_00a1995aba0f7a790169aac4b7de6c8197a73e9086863bb349"
is_byok
false
latency
180
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-1772799159-xdeky13b2XGsgfki4VQ0"
upstream_id
"resp_00a1995aba0f7a790169aac4b7de6c8197a73e9086863bb349"
total_cost
0.0019775
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
51.23%
Matches word count
n/a
neededClean
false
sentences
5
wordCounts
0
19
1
19
2
19
3
18
4
19