> For the complete documentation index, see [llms.txt](https://docs.clore.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.clore.ai/guides/guides_v2-es/generacion-de-video/ai-video-generation.md).

# Generación de video con IA

Genera videos con SVD, AnimateDiff y más en Clore.ai

Genera videos usando Stable Video Diffusion, AnimateDiff y otros modelos.

{% hint style="success" %}
Todos los ejemplos se pueden ejecutar en servidores GPU alquilados a través de [Marketplace de CLORE.AI](https://clore.ai/marketplace).
{% endhint %}

## Alquilar en CLORE.AI

1. Visita [Marketplace de CLORE.AI](https://clore.ai/marketplace)
2. Filtra por tipo de GPU, VRAM y precio
3. Elige **Bajo demanda** (tarifa fija) o **Spot** (precio ofertado)
4. Configura tu pedido:
   * Selecciona la imagen de Docker
   * Configura los puertos (TCP para SSH, HTTP para interfaces web)
   * Añade variables de entorno si es necesario
   * Introduce el comando de inicio
5. Selecciona el método de pago: **CLORE**, **BTC**, o **USDT/USDC**
6. Crea el pedido y espera al despliegue

### Accede a tu servidor

* Encuentra los detalles de conexión en **Mis pedidos**
* Interfaces web: Usa la URL del puerto HTTP
* SSH: `ssh -p <port> root@<proxy-address>`

## Modelos disponibles

| Modelo      | Tipo           | VRAM  | Duración     |
| ----------- | -------------- | ----- | ------------ |
| SVD         | Imagen a video | 16GB  | 4 segundos   |
| SVD-XT      | Imagen a video | 20 GB | 4 segundos   |
| AnimateDiff | Texto a video  | 12GB  | 2-4 segundos |
| CogVideoX   | Texto a video  | 24GB  | 6 segundos   |

## Stable Video Diffusion (SVD)

### Despliegue rápido

**Imagen de Docker:**

```
pytorch/pytorch:2.11.0-cuda12.8-cudnn9-devel
```

**Puertos:**

```
22/tcp
7860/http
```

**Comando:**

```bash
pip install diffusers transformers accelerate gradio imageio && \\
python svd_server.py
```

## Accediendo a tu servicio

Después del despliegue, encuentra tu `http_pub` URL en **Mis pedidos**:

1. Ve a **Mis pedidos** página
2. Haz clic en tu pedido
3. Encuentra la `http_pub` URL (p. ej., `abc123.clorecloud.net`)

Usa `https://YOUR_HTTP_PUB_URL` en lugar de `localhost` en los ejemplos a continuación.

### Script de SVD

```python
import torch
from diffusers import StableVideoDiffusionPipeline
from PIL import Image
import imageio

# Cargar modelo
pipe = StableVideoDiffusionPipeline.from_pretrained(
    "stabilityai/stable-video-diffusion-img2vid-xt",
    torch_dtype=torch.float16,
)
pipe.to("cuda")
pipe.enable_model_cpu_offload()

# Cargar y redimensionar imagen
image = Image.open("input.png").resize((1024, 576))

# Generar video
frames = pipe(
    image,
    decode_chunk_size=8,
    num_frames=25,
    motion_bucket_id=127,
    noise_aug_strength=0.02
).frames[0]

# Guardar como GIF
imageio.mimsave("output.gif", frames, fps=6)

# Guardar como MP4
imageio.mimsave("output.mp4", frames, fps=6)
```

### SVD con interfaz de Gradio

```python
import gradio as gr
import torch
from diffusers import StableVideoDiffusionPipeline
from PIL import Image
import imageio
import tempfile

pipe = StableVideoDiffusionPipeline.from_pretrained(
    "stabilityai/stable-video-diffusion-img2vid-xt",
    torch_dtype=torch.float16,
)
pipe.enable_model_cpu_offload()

def generate_video(image, motion_bucket, fps, num_frames):
    image = image.resize((1024, 576))

    frames = pipe(
        image,
        decode_chunk_size=4,
        num_frames=num_frames,
        motion_bucket_id=motion_bucket,
    ).frames[0]

    with tempfile.NamedTemporaryFile(suffix=".mp4", delete=False) as f:
        imageio.mimsave(f.name, frames, fps=fps)
        return f.name

demo = gr.Interface(
    fn=generate_video,
    inputs=[
        gr.Image(type="pil", label="Imagen de entrada"),
        gr.Slider(1, 255, value=127, label="Cantidad de movimiento"),
        gr.Slider(1, 30, value=6, label="FPS"),
        gr.Slider(14, 25, value=25, label="Fotogramas")
    ],
    outputs=gr.Video(label="Video generado"),
)

demo.launch(server_name="0.0.0.0", server_port=7860)
```

## AnimateDiff

### Instalación

```bash
pip install diffusers transformers accelerate
```

### Generar video a partir de texto

```python
import torch
from diffusers import AnimateDiffPipeline, MotionAdapter, DDIMScheduler
import imageio

# Cargar adaptador de movimiento
adapter = MotionAdapter.from_pretrained("guoyww/animatediff-motion-adapter-v1-5-2")

# Cargar pipeline
pipe = AnimateDiffPipeline.from_pretrained(
    "runwayml/stable-diffusion-v1-5",
    motion_adapter=adapter,
    torch_dtype=torch.float16,
)
pipe.scheduler = DDIMScheduler.from_pretrained(
    "runwayml/stable-diffusion-v1-5",
    subfolder="scheduler",
    clip_sample=False,
    timestep_spacing="linspace",
    beta_schedule="linear",
    steps_offset=1,
)
pipe.to("cuda")
pipe.enable_model_cpu_offload()

# Generar
output = pipe(
    prompt="Un gato caminando por un jardín, hermosas flores, día soleado",
    negative_prompt="mala calidad, borroso",
    num_frames=16,
    guidance_scale=7.5,
    num_inference_steps=25,
)

# Guardar
frames = output.frames[0]
imageio.mimsave("animatediff.gif", frames, fps=8)
```

### AnimateDiff con modelo personalizado

```python
from diffusers import AnimateDiffPipeline, MotionAdapter, EulerDiscreteScheduler

adapter = MotionAdapter.from_pretrained("guoyww/animatediff-motion-adapter-v1-5-2")

# Usar un checkpoint personalizado (p. ej., RealisticVision)
pipe = AnimateDiffPipeline.from_pretrained(
    "SG161222/Realistic_Vision_V5.1_noVAE",
    motion_adapter=adapter,
    torch_dtype=torch.float16,
)
```

## AnimateDiff en ComfyUI

### Instalar nodos

```bash
cd /workspace/ComfyUI/custom_nodes
git clone https://github.com/Kosinkadink/ComfyUI-AnimateDiff-Evolved.git
git clone https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite.git
```

### Descargar modelos de movimiento

```bash
cd /workspace/ComfyUI/custom_nodes/ComfyUI-AnimateDiff-Evolved/models
wget https://huggingface.co/guoyww/animatediff/resolve/main/mm_sd_v15_v2.ckpt
```

## CogVideoX

### Texto a video

```python
import torch
from diffusers import CogVideoXPipeline
import imageio

pipe = CogVideoXPipeline.from_pretrained(
    "THUDM/CogVideoX-2b",
    torch_dtype=torch.float16
)
pipe.to("cuda")
pipe.enable_model_cpu_offload()

prompt = "Un dron volando sobre un hermoso paisaje montañoso al atardecer"

video = pipe(
    prompt=prompt,
    num_videos_per_prompt=1,
    num_inference_steps=50,
    num_frames=49,
    guidance_scale=6,
).frames[0]

imageio.mimsave("cogvideo.mp4", video, fps=8)
```

## Escalado de video

### Real-ESRGAN para video

```python
import cv2
import torch
from basicsr.archs.rrdbnet_arch import RRDBNet
from realesrgan import RealESRGANer

model = RRDBNet(num_in_ch=3, num_out_ch=3, num_feat=64, num_block=23, num_grow_ch=32, scale=4)
upsampler = RealESRGANer(
    scale=4,
    model_path='RealESRGAN_x4plus.pth',
    model=model,
    tile=400,
    tile_pad=10,
    pre_pad=0,
    half=True
)

# Procesar video fotograma por fotograma
cap = cv2.VideoCapture("input.mp4")

# ... escalar cada fotograma
```

## Interpolación (videos fluidos)

### Interpolación de fotogramas FILM

```python

# Instalar
pip install tensorflow tensorflow_hub

import tensorflow as tf
import tensorflow_hub as hub

model = hub.load("https://tfhub.dev/google/film/1")

def interpolate(frame1, frame2, num_interpolations=3):
    # Devuelve fotogramas interpolados entre frame1 y frame2
    ...
```

### RIFE (tiempo real)

```bash
pip install rife-ncnn-vulkan-python

from rife_ncnn_vulkan import Rife
rife = Rife(gpu_id=0)

# Interpolar fotogramas
```

## Generación por lotes de video

```python
prompts = [
    "Un cohete despegando hacia el espacio",
    "Las olas del océano rompiendo contra las rocas",
    "Una mariposa volando entre flores",
]

for i, prompt in enumerate(prompts):
    print(f"Generando {i+1}/{len(prompts)}")
    output = pipe(prompt, num_frames=16)
    imageio.mimsave(f"video_{i:03d}.mp4", output.frames[0], fps=8)
```

## Consejos de memoria

### Para VRAM limitada

```python

# Habilitar descarga a CPU
pipe.enable_model_cpu_offload()

# Habilitar segmentación de VAE
pipe.enable_vae_slicing()

# Habilitar segmentación de la atención
pipe.enable_attention_slicing()

# Reducir el número de fotogramas
num_frames = 14  # En lugar de 25
```

### Decodificación por bloques

```python
frames = pipe(
    image,
    decode_chunk_size=2,  # Decodificar 2 fotogramas a la vez
    num_frames=25,
).frames[0]
```

## Conversión de salida

### GIF a MP4

```bash
ffmpeg -i input.gif -movflags faststart -pix_fmt yuv420p -vf "scale=trunc(iw/2)*2:trunc(ih/2)*2" output.mp4
```

### Secuencia de fotogramas a video

```bash
ffmpeg -framerate 8 -i frame_%04d.png -c:v libx264 -pix_fmt yuv420p output.mp4
```

### Añadir audio

```bash
ffmpeg -i video.mp4 -i audio.mp3 -c:v copy -c:a aac -shortest output_with_audio.mp4
```

## Rendimiento

| Modelo      | GPU      | Fotogramas | Tiempo  |
| ----------- | -------- | ---------- | ------- |
| SVD-XT      | RTX 3090 | 25         | \~120 s |
| SVD-XT      | RTX 4090 | 25         | \~80 s  |
| SVD-XT      | A100     | 25         | \~50 s  |
| AnimateDiff | RTX 3090 | 16         | \~30 s  |
| CogVideoX   | A100     | 49         | \~180 s |

## Estimación de costos

Tarifas típicas del marketplace de CLORE.AI (a partir de 2024):

| GPU       | Tarifa por hora | Tarifa diaria | Sesión de 4 horas |
| --------- | --------------- | ------------- | ----------------- |
| RTX 3060  | \~$0.03         | \~$0.70       | \~$0.12           |
| RTX 3090  | \~$0.06         | \~$1.50       | \~$0.25           |
| RTX 4090  | \~$0.10         | \~$2.30       | \~$0.40           |
| A100 40GB | \~$0.17         | \~$4.00       | \~$0.70           |
| A100 80GB | \~$0.25         | \~$6.00       | \~$1.00           |

*Los precios varían según el proveedor y la demanda. Consulta* [*Marketplace de CLORE.AI*](https://clore.ai/marketplace) *las tarifas actuales.*

**Ahorra dinero:**

* Usa el **Spot** mercado para trabajo interrumpible — alrededor de un tercio de los servidores fija el precio spot por debajo del precio bajo demanda (mediana de \~13% de descuento), el resto lo iguala
* Paga con **CLORE** tokens
* Compara precios entre distintos proveedores

## Solución de problemas

### Error de OOM

* Reduce num\_frames
* Habilita la descarga a CPU
* Usa un decode\_chunk\_size más pequeño

### Video parpadeante

* Aumenta num\_inference\_steps
* Prueba un motion\_bucket\_id diferente
* Usa interpolación de fotogramas

### Calidad deficiente

* Usa una entrada de mayor resolución (SVD)
* Mejores prompts (AnimateDiff)
* Aumenta guidance\_scale


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.clore.ai/guides/guides_v2-es/generacion-de-video/ai-video-generation.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.