NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-02-08-01KGZ019GDNQZM7ZQSE253FME4
Run Details
Model:
Claude Sonnet 4.5
Test:
N-Length Sentences
Scenario:
Write sentences with 10 words each
Duration:
5.5s
Cost:
$0.003087
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
98.15%
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
The body lay crumpled beneath the grand marble library staircase. Detective Sarah Chen arrived precisely at dawn, notebook in hand. Rain had washed away potential evidence from the front steps. The victim wore an expensive tuxedo, blood staining white fabric. Mrs. Hawthorne discovered him during her usual morning cleaning routine. She screamed loud enough to wake the entire sleeping household. Six guests had stayed overnight after the charity gala ended. Each one claimed innocence, each one had secrets to hide. The murder weapon, a silver letter opener, lay nearby. Sarah photographed the scene before carefully examining the victim's pockets. Inside she found a crumpled note with ten words written. "Meet me in the library at midnight. Come alone tonight." Sarah looked up at the nervous faces surrounding the scene. One of them was lying, one of them was guilty.
created_at
"2026-02-08T16:06:01.94494+00:00"
2/8/2026, 4:06:01 PM
model
"anthropic/claude-4.5-sonnet-20250929"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1367
moderation_latency
169
generation_time
5288
tokens_prompt
35
tokens_completion
170
native_tokens_prompt
44
native_tokens_completion
197
native_tokens_completion_images
(null)
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
"stop"
usage
0.003087
router
(null)
provider_responses
0
id
"db762397-c394-4e2a-b96a-c3a78cb9b3a3"
status
200
is_byok
false
latency
1367
endpoint_id
"d8bee848-f020-454a-b2be-4a7ce868f99f"
provider_name
"Amazon Bedrock"
model_permaslug
"anthropic/claude-4.5-sonnet-20250929"
api_type
"completions"
id
"gen-1770566755-BpG95OFMwU218JT4e4YW"
upstream_id
"db762397-c394-4e2a-b96a-c3a78cb9b3a3"
total_cost
0.003087
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
Evaluation details
Result
Evaluator
Details
Meta Data
98.15%
Matches word count
n/a
neededClean
false
sentences
14
wordCounts
0
10
1
10
2
10
3
10
4
10
5
10
6
10
7
10
8
9
9
10
10
10
11
10
12
10
13
10