NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5SEKFGVKAYFNYVTKR7J6B
Run Details
Model:
DeepSeek V3.1
Test:
N-Length Sentences
Scenario:
Write sentences with 5 words each
Duration:
5.4s
Cost:
$0.000164
Provider:
AtlasCloud
Model input
User/Human
Please write the opening scene of a murder mystery where each sentence has exactly 5 words. Please make sure that each sentence has the same length. Only return the opening itself.
Model response
Messages
Meta
AI
The body lay in darkness. A single lamp flickered overhead. It cast long, dancing shadows. A man was quite dead. He wore a formal black suit. A silver dagger gleamed brightly. It pierced his silent heart. No other wound was visible. The room was perfectly still. A grandfather clock ticked softly. It broke the deep silence. Detective Miles arrived at the scene. He observed the meticulous arrangement. Everything seemed too perfectly placed. This was no crime passion. This murder was a statement. He knelt by the corpse. A faint scent of roses. It clung to the air. The victim’s hand was open. A small, torn photograph remained. It showed a grinning face. But the face was ripped. Only the smile remained intact. Someone had left a message.
created_at
"2026-02-16T21:30:39.811671+00:00"
2/16/2026, 9:30:39 PM
model
"deepseek/deepseek-chat-v3.1"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1680
moderation_latency
(null)
generation_time
5316
tokens_prompt
65
tokens_completion
187
native_tokens_prompt
40
native_tokens_completion
160
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
0.000164
router
(null)
provider_responses
0
id
"f740eb86c4cf4020a3943403644c8bfe"
status
200
is_byok
false
latency
1678
endpoint_id
"0bfe3535-ed02-47ea-a98f-60cc8af4fab2"
provider_name
"AtlasCloud"
model_permaslug
"deepseek/deepseek-chat-v3.1"
api_type
"completions"
id
"gen-1771277433-yBXjk2eREtSvzDkTtU4U"
upstream_id
"f740eb86c4cf4020a3943403644c8bfe"
total_cost
0.000164
cache_discount
(null)
upstream_inference_cost
0
provider_name
"AtlasCloud"
Evaluation details
Result
Evaluator
Details
Meta Data
98.2304%
Matches word count
n/a
neededClean
false
sentences
25
wordCounts
0
5
1
5
2
5
3
5
4
6
5
5
6
5
7
5
8
5
9
5
10
5
11
6
12
5
13
5
14
5
15
5
16
5
17
5
18
5
19
5
20
5
21
5
22
5
23
5
24
5