NC Bench
Overview
About
Tests
Models
Model
Gemini 2.5 Pro
Test
N-Length Sentences
Scenario
Write sentences with 20 words each
Run ID
2025-11-18-01KACP4QYS8Y9ME0V06GMWYWJC
Model input
User/Human
Please write the opening scene of a murder mystery where each sentence has exactly 20 words. Please make sure that each sentence has the same length. Only return the opening itself.
Model response
Messages
Meta
AI
Detective Inspector Miles Corbin surveyed the opulent room, his sharp, observant eyes missing absolutely nothing in the flickering firelight. The master of the house, Lord Alistair Harrington, was slumped lifelessly in his favorite high-backed, green leather reading chair. A single, perfect, crimson stain blossomed like a dark, ghastly rose upon the starched white front of his shirt. An overturned crystal glass lay shattered on the expensive Persian rug, its amber liquid seeping slowly into the intricate fibers. Corbin noted the strange absence of a struggle, despite the undeniable, final evidence of extremely sudden and very personal violence. In the shadowed doorway, the pale, frightened faces of the assembled family members watched his every single calculated, deliberate move.
created_at
"2025-11-18T23:52:47.90974+00:00"
Tuesday, November 18th, 2025 at 11:52:47 PM GMT+00:00
model
"google/gemini-2.5-pro"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2428
moderation_latency
(null)
generation_time
14382
tokens_prompt
42
tokens_completion
703
native_tokens_prompt
38
native_tokens_completion
1588
native_tokens_completion_images
0
native_tokens_reasoning
1440
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
0.0159275
api_type
"completions"
id
"gen-1763509952-LjjY6qzxBunifqqvXSZ7"
upstream_id
(null)
total_cost
0.0159275
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
Evaluation details
Result
Evaluator
Details
Meta Data
86.8844%
Matches word count
n/a
neededClean
false
sentences
6
wordCounts
0
19
1
20
2
19
3
20
4
20
5
20