Run: 2026-02-20-01KHY9ZZ831ZG0EYG949RQT7QY

Run Details

Model:
Gemma 3 12B
Test:
N-Length Sentences
Scenario:
Write sentences with 10 words each
Duration:
2.9s
Cost:
$0.000017
Provider:
DeepInfra

Evaluation summary

Score Evaluator
Details Link
74.08% 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

AI
Rain lashed against the windows of Blackwood Manor tonight. A grand, imposing estate, shrouded in thick, swirling mist. Inside, Lord Ashworth lay sprawled upon the Persian rug. Crimson stained the opulent carpet, a horrifying, shocking sight. His loyal butler, Jeeves, discovered him just moments ago. A single, ornate letter opener protruded from his chest. Panic seized the household; everyone felt utterly, deeply afraid. Detectives arrived swiftly, eager to solve this grim puzzle. They surveyed the scene, searching for any subtle clues. The air hung heavy with secrets and unspoken fear.
created_at"2026-02-20T19:56:25.743Z" 2/20/2026, 7:56:25 PM
model"google/gemma-3-12b-it"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency398
moderation_latency(null)
generation_time2798
tokens_prompt63
tokens_completion149
native_tokens_prompt46
native_tokens_completion120
native_tokens_completion_images(null)
native_tokens_reasoning0
native_tokens_cached0
num_media_prompt(null)
num_input_audio_prompt(null)
num_media_completion0
num_search_results(null)
origin"https://ncbench.com/"
is_byokfalse
finish_reason"stop"
native_finish_reason"stop"
usage0.00001744
router(null)
provider_responses
0
endpoint_id"eb06dc92-5a16-47ec-a776-6ef956457c47"
id"chatcmpl-RRmfuTRsXvjajFlvICqc5liU"
is_byokfalse
latency88
model_permaslug"google/gemma-3-12b-it"
provider_name"DeepInfra"
status200
api_type"completions"
id"gen-1771617385-ABgtxvNgvaY0SGvbQJb5"
upstream_id"chatcmpl-RRmfuTRsXvjajFlvICqc5liU"
total_cost0.00001744
cache_discount(null)
upstream_inference_cost0
provider_name"DeepInfra"

Evaluation details

Result Evaluator Details Meta Data
74.08% Matches word count n/a
neededCleanfalse
sentences10
wordCounts
09
19
29
39
49
59
69
79
89
99