NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY8QDJANXGBHZ6WBXH75SHS
Run Details
Model:
Gemma 3 4B
Test:
N-Length Sentences
Scenario:
Write sentences with 10 words each
Duration:
2.2s
Cost:
$0.000012
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
72.28%
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
Rain lashed against the windows, a mournful, dismal scene. Detective Harding surveyed the opulent study, utterly serene. Crimson stained the Persian rug, a horrifying, brutal sight. The victim lay sprawled lifelessly, beneath a priceless, dark art. A single gunshot echoed faintly, unsettlingly cold and deep. Guests whispered nervously, avoiding glances at the gruesome display. Inspector Davies examined footprints, carefully documenting each dusty trace. The butler stood frozen, holding a silver, ornate tray. Something felt terribly wrong, a disturbing, unsettling secret known. Harding knew this investigation would prove intensely, tragically slow.
created_at
"2026-02-20T19:34:16.922Z"
2/20/2026, 7:34:16 PM
model
"google/gemma-3-4b-it"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
163
moderation_latency
(null)
generation_time
2087
tokens_prompt
63
tokens_completion
165
native_tokens_prompt
46
native_tokens_completion
126
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
0.00001192
router
(null)
provider_responses
0
endpoint_id
"d3de3bd4-81bc-48fb-924f-2a87b2a36e75"
id
"chatcmpl-RJZiWxkwm0yQnrPjlwZ94Ire"
is_byok
false
latency
57
model_permaslug
"google/gemma-3-4b-it"
provider_name
"DeepInfra"
status
200
api_type
"completions"
id
"gen-1771616056-O5IsvZQ9aZ618AesMs6n"
upstream_id
"chatcmpl-RJZiWxkwm0yQnrPjlwZ94Ire"
total_cost
0.00001192
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
72.28%
Matches word count
n/a
neededClean
false
sentences
10
wordCounts
0
9
1
8
2
9
3
10
4
9
5
9
6
9
7
9
8
9
9
9