diff --git a/built-in-nodes/loaders/checkpoint_loader.mdx b/built-in-nodes/loaders/checkpoint_loader.mdx
index a4ce85e..85c466d 100644
--- a/built-in-nodes/loaders/checkpoint_loader.mdx
+++ b/built-in-nodes/loaders/checkpoint_loader.mdx
@@ -1,32 +1,13 @@
----
-title: "Load Checkpoint"
----
+Create a high-quality AI-generated short animation based on the following text prompt:
 
-The Load Checkpoint node can be used to load a diffusion model, diffusion models are used to denoise latents. This node will also provide the appropriate VAE and CLIP model.
+"A mysterious cyberpunk girl walks through a neon-lit alley at night, rain pouring down, with holograms flickering around her. She slowly turns to face the camera as a glowing drone hovers beside her."
 
-## Inputs 
+- Style: Cinematic, highly detailed, anime-inspired visuals
+- Format: 16:9 landscape, 768x432 resolution
+- Duration: 4–6 seconds
+- Mood: Moody, dramatic, futuristic
+- Content: This is a SFW version. NSFW variants should follow the same scene layout but with adult-oriented detail, respecting platform content rules.
 
-<ResponseField name="ckpt_name">
+Use AnimateDiff with SDXL models or custom fine-tuned NSFW models. Add subtle camera motion, breathing animation, and smooth transitions between frames.
 
-
-The name of the model.
-</ResponseField>
-
-
-## Outputs
-
-<ResponseField name="MODEL">
-
-The model used for denoising latents.
-
-</ResponseField>
-
-<ResponseField name="CLIP">
-
-The CLIP model used for encoding text prompts.
-</ResponseField>
-
-<ResponseField name="VAE">
-
-The VAE model used for encoding and decoding images to and from latent space.
-</ResponseField>
\ No newline at end of file
+Generate the keyframes using ComfyUI or AUTOMATIC1111 + ControlNet to guide motion (e.g., walking, head turns). Use a consistent seed for character stability. Finally, export the animation at 24fps using FFMPEG or Video Enhance AI for upscaling if needed.