NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-09-01KKAC3KH4C6ECEBG5DQPBN3ZN
Run Details
Model:
GPT-5.4 (Reasoning, Low)
Test:
N-Length Sentences
Scenario:
Write sentences with 10 words each
Duration:
8.9s
Cost:
$0.007197
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
Model input
User/Human
Please write the opening scene of a murder mystery where each sentence has exactly 10 words. Please make sure that each sentence has the same length. Only return the opening itself.
Model response
Messages
Meta
AI
Dawn leaked through stained glass, painting corpses with holy colors. Inspector Vale arrived early, smelling rain, tobacco, and old regret. The cathedral stood silent, except for flies worrying the altar. Beneath the crucifix, Lady Harrow knelt, throat opened like confession. Three candles flickered nearby, though none remembered lighting them tonight. Her pearl rosary had snapped, scattering moons across the flagstones. Vale noticed muddy footprints leading inward, but never coming back. Then the organ groaned once, and someone laughed behind him.
created_at
"2026-03-09T22:39:56.974Z"
3/9/2026, 10:39:56 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
7620
moderation_latency
(null)
generation_time
8849
tokens_prompt
46
tokens_completion
271
native_tokens_prompt
41
native_tokens_completion
473
native_tokens_completion_images
0
native_tokens_reasoning
357
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
0.0071975
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_08d96377a1679f840169af4c3d0c188195beeb9a1534965422"
is_byok
false
latency
221
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-1773095996-3FdEU6Ye0t77xWfAcGvW"
upstream_id
"resp_08d96377a1679f840169af4c3d0c188195beeb9a1534965422"
total_cost
0.0071975
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
sentences
8
wordCounts
0
10
1
10
2
10
3
10
4
10
5
10
6
10
7
10