| --- |
| license: apache-2.0 |
| language: |
| - en |
| - zh |
| - th |
| base_model: text-to-video-synthesis |
| pipeline_tag: image-to-video |
| library_name: gguf |
| tags: |
| - image-to-video |
| - wan2.2 |
| - comfyui |
| - GGUF |
| --- |
| |
| # Wan2.2 Custom GGUF (Tesla T4 Optimized) |
|
|
| This repository provides highly optimized **Wan2.2 Image-to-Video (I2V) GGUF+LIGHNINGV2** and custom models. These variants are fine-tuned for running efficiently on memory-constrained environments, such as Google Colab equipped with an NVIDIA Tesla T4 GPU. |
|
|
| --- |
|
|
| ## ⚡ Optimal Settings for ComfyUI |
|
|
| To achieve perfect video motion without artifacts or image degradation (preventing fried or oversaturated visuals), we strongly recommend using the following parameters: |
|
|
|
|
| | Parameter | Recommended Value | Note | |
| | :--- | :--- | :--- | |
| | **Sampling Steps** | `4` | When using Wan2.2 Lightning / Distilled V2 | |
| | **CFG Scale** | `1.0` | Crucial for preventing burnt images | |
| | **High Noise Steps** | `2` or `3` | To lock in strong motion and structure before the Lightning layer clears noise | |
| | **Low Noise Steps** | `3` (End Step: `4`) | Fine-tuning phase | |
| | **Sampler / Scheduler** | `euler` + `simple` | Standard diffusion setup | |
|
|
| ### 👑 Note for Higher Quality (Hybrid Workflow & Hardware Restrictions): |
| If you want to achieve higher visual fidelity and enhance micro-details, using **`wan2.2_i2v_low_noise_14B_fp8_scaled.safetensors`** as your final step is highly recommended. This hybrid approach significantly sharpens fine details and effectively eliminates motion blur. |
|
|
| However, you **MUST** follow these hardware restrictions based on your execution environment to avoid crashes: |
|
|
| * 🚀 **Via Backdoor (Direct Code / Colab Forms):** You can run the `fp8_scaled` model even on a **Tesla T4 (Free Tier)**, but you *MUST properly configure and initialize your memory management system* in the script background before execution. |
| * 💻 **Via ComfyUI GUI (Web Interface):** |
| * **NVIDIA L4 (24GB VRAM) or higher** is strictly recommended due to the heavy web GUI memory overhead. |
| * **⚠️ CRITICAL WARNING FOR T4 GUI:** If you are running via the ComfyUI Web GUI on an **NVIDIA Tesla T4**, **DO NOT use the `fp8_scaled` model!** It will cause an immediate **OOM (Out of Memory) crash**. Instead, you must use the hybrid setup: **`Q6_K` (High Noise) + `Q8_H` (Low Noise)**, which is perfectly safe and highly optimized for T4 GUI limits. |
| |
| --- |
| |
| ## 💾 Available Model Variants |
| |
| Choose the right variant based on your creative workflow and VRAM configuration: |
| |
| ### 🔥 High Noise Models (`wan2.2_i2v_high_noise_...`) |
| * **Best for:** Creative, high-motion generation, and diverse camera movements. |
| * **Available Quantizations:** `Q4_K_M`, `Q6_K_L`, `Q6_K`, `Q8_H` |
| |
| ### ❄️ Low Noise Models (`wan2.2_i2v_low_noise_...`) |
| * **Best for:** High fidelity, generation stability, and strictly adhering to the prompt or structural layout of your starting frame. |
| * **Available Quantizations:** `Q4_K_M`, `Q6_K_L`, `Q6_K`, `Q8_H`, and `fp8_scaled` |
| |
| --- |
| |
