youtube thumbnail generation

Este extracto es una copia guardada durante la revisión. La fuente externa contiene la versión completa y actual.

---
name: YouTube Thumbnail Generation
description: Generate click-worthy YouTube thumbnails with high CTR designs using each::sense API
metadata:
  category: image-generation
  api: each::sense
  base_url: https://eachsense-agent.core.eachlabs.run/v1/chat/completions
  method: POST
  tags:
    - youtube
    - thumbnail
    - content-creation
    - social-media
    - marketing
---

# YouTube Thumbnail Generation

Generate high-converting YouTube thumbnails designed for maximum click-through rates. Create attention-grabbing visuals with expressive faces, bold colors, clear text spaces, and proven thumbnail formulas that drive views.

## Overview

The each::sense API creates YouTube thumbnails optimized for engagement:

- **High CTR Designs**: Thumbnails engineered for maximum click-through rates
- **Expressive Faces**: Shocked, excited, curious, and emotional expressions that draw attention
- **Text Space**: Clean areas for overlay text and titles
- **Bold Color Palettes**: Bright, saturated colors that stand out in feeds
- **High Contrast**: Sharp visual separation that pops at any size
- **1280x720 Resolution**: Standard YouTube thumbnail dimensions

## Quick Start

```bash
curl -X POST "https://eachsense-agent.core.eachlabs.run/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Generate a YouTube thumbnail: shocked face reaction to an unbelievable reveal, mouth wide open, eyes popping, bright yellow and red background, space on the right side for text overlay, dramatic lighting, ultra high contrast, 1280x720 aspect ratio"}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'
```

## Thumbnail Styles

| Style | Description | Key Elements |
|-------|-------------|--------------|
| **Reaction Face** | Exaggerated emotional expressions | Wide eyes, open mouth, dramatic lighting |
| **Before/After** | Side-by-side transformation | Split composition, contrast between states |
| **Listicle** | Numbered list or top 10 format | Bold numbers, multiple elements arranged |
| **Tutorial** | How-to and educational content | Steps visualization, clean layout |
| **Vlog** | Personal and lifestyle content | Candid feel, warm colors, relatable |
| **Gaming** | Game-related content | Action scenes, game elements, energetic |
| **Review** | Product or service reviews | Product focus, rating visual, comparison |

## Use Case Examples

### Reaction/Shocked Face Thumbnail

Classic YouTube thumbnail with exaggerated shocked expression.

```bash
curl -X POST "https://eachsense-agent.core.eachlabs.run/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create a YouTube thumbnail with a person showing extreme shock and disbelief, jaw dropped, eyes wide open, hands on cheeks, bright neon pink and electric blue gradient background, dramatic side lighting creating shadows, space in the upper right corner for text, hyper-saturated colors, 1280x720 YouTube thumbnail format"}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'
```

### Before/After Transformation Thumbnail

Split-screen showing dramatic transformation.

```bash
curl -X POST "https://eachsense-agent.core.eachlabs.run/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Generate a before and after YouTube thumbnail, split down the middle, left side showing a messy cluttered room in dull gray tones, right side showing the same room transformed into a beautiful organized space with warm golden lighting, red arrow pointing from left to right, high contrast, text space at top, 1280x720 aspect ratio"}],
    "model": "eachsense/beta",
    "stream