From 1cd881527a83610eaca2ea66f628d72bd3550e62 Mon Sep 17 00:00:00 2001 From: elishakay Date: Thu, 10 Jul 2025 11:52:55 +0300 Subject: [PATCH 1/7] docs upgrades --- README.md | 136 +++++------------------------------------- docs/release-notes.md | 123 ++++++++++++++++++++++++++++++++++++++ 2 files changed, 137 insertions(+), 122 deletions(-) create mode 100644 docs/release-notes.md diff --git a/README.md b/README.md index 0f1c335..ccb0f9f 100644 --- a/README.md +++ b/README.md @@ -1,3 +1,5 @@ +
+ # ComfyUI-LTXVideo [![GitHub](https://img.shields.io/badge/LTXV-Repo-blue?logo=github)](https://github.com/Lightricks/LTX-Video) @@ -8,135 +10,25 @@ [![Paper](https://img.shields.io/badge/Paper-arXiv-B31B1B?logo=arxiv)](https://arxiv.org/abs/2501.00103) [![Discord](https://img.shields.io/badge/Join-Discord-5865F2?logo=discord)](https://discord.gg/Mn8BRgUKKy) -ComfyUI-LTXVideo is a collection of custom nodes for ComfyUI, designed to provide useful tools for working with the LTXV model. -The model itself is supported in the core ComfyUI [code](https://github.com/comfyanonymous/ComfyUI/tree/master/comfy/ldm/lightricks). -The main LTXVideo repository can be found [here](https://github.com/Lightricks/LTX-Video). - -# ⭐ 8.07.2025 - LTXVideo ICLora Release ⭐ -### πŸš€ What's New in LTXVideo ICLoRA - -1. **Three New ICLoRA Models** - Introducing powerful in-context LoRA models that enable precise control over video generation: - - Depth Control: [LTX-Video-ICLoRA-depth-13b-0.9.7](https://huggingface.co/Lightricks/LTX-Video-ICLoRA-depth-13b-0.9.7) - - Pose Control: [LTX-Video-ICLoRA-pose-13b-0.9.7](https://huggingface.co/Lightricks/LTX-Video-ICLoRA-pose-13b-0.9.7) - - Edge/Canny Control: [LTX-Video-ICLoRA-canny-13b-0.9.7](https://huggingface.co/Lightricks/LTX-Video-ICLoRA-canny-13b-0.9.7) - -2. **New Node: πŸ…›πŸ…£πŸ…§ LTXV In Context Sampler** - A dedicated node for seamlessly integrating ICLoRA models into your workflow, enabling fine-grained control over video generation using depth maps, pose estimation, or edge detection. - -3. **Example Workflow** - Check out [example workflow](#iclora) for a complete example demonstrating how to use the ICLoRA models effectively. - -4. **Custom ICLoRA Training** - We've released a trainer that allows you to create your own specialized ICLoRA models for custom control signals. Check out the [trainer repository](https://github.com/Lightricks/LTX-Video-Trainer) to get started. - - -# ⭐ 9.06.2025 – LTXVideo VAE Patcher, Mask manipulation and Q8 LoRA loader nodes. ⭐ -1. **LTXV Patcher VAE**
The new node improves VAE decoding performance by reducing runtime and cutting memory consumption by up to 50%. This allows generation of higher-resolution outputs on consumer-grade GPUs with limited VRAM, without needing to load the VAE partially or decode in tiles.
-⚠️ On *Windows*, you may need to add the paths to the *MSVC compiler (cl.exe)* and *ninja.exe* to your system environment PATH variable.
-2. **LTXV Preprocess Masks**
-Preprocesses masks for use with the LTXVideo model's latent masking. It validates mask dimensions based on VAE downscaling, supports optional inversion, handles the first frame mask separately, combines temporal masks via max pooling, applies morphological operations to grow or shrink masks, and clamps values to ensure correct opacity. The result is a set of masks ready for latent-space masking. -3. **LTXV Q8 Lora Model Loader**
-Applying LoRA to an FP8-quantized model requires special handling to preserve output quality. It's crucial to apply LoRA weights using the correct precision, as the current LoRA implementation in ComfyUI does so in a non-optimal way. This node addresses that limitation by ensuring LoRA weights are applied properly, resulting in significantly better quality. If you're working with an FP8 LTXV model, using this node guarantees that LoRA behaves as expected and delivers the intended effect. - - -# ⭐ 14.05.2025 – LTXVideo 13B 0.9.7 Distilled Release ⭐ - -### πŸš€ What's New in LTXVideo 13B 0.9.7 Distilled -1. **LTXV 13B Distilled πŸ₯³ 0.9.7**
- Delivers cinematic-quality videos at fraction of steps needed to run full model. Only 4 or 8 steps needed for single generation.
- πŸ‘‰ [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-13b-0.9.7-distilled.safetensors) - -2. **LTXV 13B Distilled Quantized 0.9.7**
- Offers reduced memory requirements and even faster inference speeds. - Ideal for consumer-grade GPUs (e.g., NVIDIA 4090, 5090).
- ***Important:*** In order to get the best performance with the quantized version please install [q8_kernels](https://github.com/Lightricks/LTXVideo-Q8-Kernels) package and use dedicated flow below.
- πŸ‘‰ [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-13b-0.9.7-distilled-fp8.safetensors)
- 🧩 Example ComfyUI flow available in the [Example Workflows](#example-workflows) section. - -3. **Updated LTV 13B Quantized version**
-From now on all our 8bit quantized models are running natively in ComfyUI, still with our Q8 patcher node you will get the best inference speed.
-πŸ‘‰ [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-13b-0.9.7-dev-fp8.safetensors)
-# ⭐ 06.05.2025 – LTXVideo 13B 0.9.7 Release ⭐ - -### πŸš€ What's New in LTXVideo 13B 0.9.7 - -1. **LTXV 13B 0.9.7** - Delivers cinematic-quality videos at unprecedented speed.
- πŸ‘‰ [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-13b-0.9.7-dev.safetensors) +
-2. **LTXV 13B Quantized 0.9.7** - Offers reduced memory requirements and even faster inference speeds. - Ideal for consumer-grade GPUs (e.g., NVIDIA 4090, 5090). - Delivers outstanding quality with improved performance.
- ***Important:*** In order to run the quantized version please install [LTXVideo-Q8-Kernels](https://github.com/Lightricks/LTXVideo-Q8-Kernels) package and use dedicated flow below. Loading the model in Comfy with LoadCheckpoint node won't work.
- πŸ‘‰ [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-13b-0.9.7-dev-fp8.safetensors)
- 🧩 Example ComfyUI flow available in the [Example Workflows](#example-workflows) section. -3. **Latent Upscaling Models** - Enables inference across multiple scales by upscaling latent tensors without decoding/encoding. - Multiscale inference delivers high-quality results in a fraction of the time compared to similar models.
- ***Important:*** Make sure you put the models below in **models/upscale_models** folder.
- πŸ‘‰ Spatial upscaling: [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-spatial-upscaler-0.9.7.safetensors).
- πŸ‘‰ Temporal upscaling: [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-temporal-upscaler-0.9.7.safetensors).
- 🧩 Example ComfyUI flow available in the [Example Workflows](#example-workflows) section. +## Table of Contents +- [Introduction](#introduction) +- [Installation](#installation) +- [Workflows](#installation) +- [What's new](./docs/release-notes.md) -### Technical Updates -1. ***New simplified flows and nodes***
-1.1. Simplified image to video: [Download here](example_workflows/ltxv-13b-i2v-base.json).
-1.2. Simplified image to video with extension: [Download here](example_workflows/ltxv-13b-i2v-extend.json).
-1.3. Simplified image to video with keyframes: [Download here](example_workflows/ltxv-13b-i2v-keyframes.json).
+# Introduction -# 17.04.2025 ⭐ LTXVideo 0.9.6 Release ⭐ - -### LTXVideo 0.9.6 introduces: - -1. LTXV 0.9.6 – higher quality, faster, great for final output. Download from [here](https://huggingface.co/Lightricks/LTX-Video/resolve/main/ltxv-2b-0.9.6-dev-04-25.safetensors). -2. LTXV 0.9.6 Distilled – our fastest model yet (only 8 steps for generation), lighter, great for rapid iteration. Download from [here](https://huggingface.co/Lightricks/LTX-Video/resolve/main/ltxv-2b-0.9.6-distilled-04-25.safetensors). - -### Technical Updates - -We introduce the __STGGuiderAdvanced__ node, which applies different CFG and STG parameters at various diffusion steps. All flows have been updated to use this node and are designed to provide optimal parameters for the best quality. -See the [Example Workflows](#example-workflows) section. - -# 5.03.2025 ⭐ LTXVideo 0.9.5 Release ⭐ - -### LTXVideo 0.9.5 introduces: - -1. Improved quality with reduced artifacts. -2. Support for higher resolution and longer sequences. -3. Frame and sequence conditioning (beyond the first frame). -4. Enhanced prompt understanding. -5. Commercial license availability. - -### Technical Updates - -Since LTXVideo is now fully supported in the ComfyUI core, we have removed the custom model implementation. Instead, we provide updated workflows to showcase the new features: - -1. **Frame Conditioning** – Enables interpolation between given frames. -2. **Sequence Conditioning** – Allows motion interpolation from a given frame sequence, enabling video extension from the beginning, end, or middle of the original video. -3. **Prompt Enhancer** – A new node that helps generate prompts optimized for the best model performance. - See the [Example Workflows](#example-workflows) section for more details. - -### LTXTricks Update - -The LTXTricks code has been integrated into this repository (in the `/tricks` folder) and will be maintained here. The original [repo](https://github.com/logtd/ComfyUI-LTXTricks) is no longer maintained, but all existing workflows should continue to function as expected. - -## 22.12.2024 - -Fixed a bug which caused the model to produce artifacts on short negative prompts when using a native CLIP Loader node. - -## 19.12.2024 ⭐ Update ⭐ +ComfyUI-LTXVideo is a collection of custom nodes for ComfyUI, designed to provide useful tools for working with the LTXV model. +The model itself is supported in the core ComfyUI [code](https://github.com/comfyanonymous/ComfyUI/tree/master/comfy/ldm/lightricks). +The main LTXVideo repository can be found [here](https://github.com/Lightricks/LTX-Video). -1. Improved model - removes "strobing texture" artifacts and generates better motion. Download from [here](https://huggingface.co/Lightricks/LTX-Video/resolve/main/ltx-video-2b-v0.9.1.safetensors). -2. STG support -3. Integrated image degradation system for improved motion generation. -4. Additional initial latent optional input to chain latents for high res generation. -5. Image captioning in image to video [flow](example_workflows/ltxvideo-i2v.json). -## Installation +# Installation Installation via [ComfyUI-Manager](https://github.com/ltdrdata/ComfyUI-Manager) is preferred. Simply search for `ComfyUI-LTXVideo` in the list of nodes and follow installation instructions. @@ -161,7 +53,7 @@ For portable ComfyUI installations, run 1. Download [ltx-video-2b-v0.9.1.safetensors](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltx-video-2b-v0.9.1.safetensors) from Hugging Face and place it under `models/checkpoints`. 2. Install one of the t5 text encoders, for example [google_t5-v1_1-xxl_encoderonly](https://huggingface.co/mcmonkey/google_t5-v1_1-xxl_encoderonly/tree/main). You can install it using ComfyUI Model Manager. -## Example workflows +# Workflows Note that to run the example workflows, you need to have some additional custom nodes, like [ComfyUI-VideoHelperSuite](https://github.com/kosinkadink/ComfyUI-VideoHelperSuite) and others, installed. You can do it by pressing "Install Missing Custom Nodes" button in ComfyUI Manager. diff --git a/docs/release-notes.md b/docs/release-notes.md new file mode 100644 index 0000000..24032c5 --- /dev/null +++ b/docs/release-notes.md @@ -0,0 +1,123 @@ +# ⭐ 8.07.2025 - LTXVideo ICLora Release ⭐ +### πŸš€ What's New in LTXVideo ICLoRA + +1. **Three New ICLoRA Models** + Introducing powerful in-context LoRA models that enable precise control over video generation: + - Depth Control: [LTX-Video-ICLoRA-depth-13b-0.9.7](https://huggingface.co/Lightricks/LTX-Video-ICLoRA-depth-13b-0.9.7) + - Pose Control: [LTX-Video-ICLoRA-pose-13b-0.9.7](https://huggingface.co/Lightricks/LTX-Video-ICLoRA-pose-13b-0.9.7) + - Edge/Canny Control: [LTX-Video-ICLoRA-canny-13b-0.9.7](https://huggingface.co/Lightricks/LTX-Video-ICLoRA-canny-13b-0.9.7) + +2. **New Node: πŸ…›πŸ…£πŸ…§ LTXV In Context Sampler** + A dedicated node for seamlessly integrating ICLoRA models into your workflow, enabling fine-grained control over video generation using depth maps, pose estimation, or edge detection. + +3. **Example Workflow** + Check out [example workflow](#iclora) for a complete example demonstrating how to use the ICLoRA models effectively. + +4. **Custom ICLoRA Training** + We've released a trainer that allows you to create your own specialized ICLoRA models for custom control signals. Check out the [trainer repository](https://github.com/Lightricks/LTX-Video-Trainer) to get started. + + +# ⭐ 9.06.2025 – LTXVideo VAE Patcher, Mask manipulation and Q8 LoRA loader nodes. ⭐ +1. **LTXV Patcher VAE**
The new node improves VAE decoding performance by reducing runtime and cutting memory consumption by up to 50%. This allows generation of higher-resolution outputs on consumer-grade GPUs with limited VRAM, without needing to load the VAE partially or decode in tiles.
+⚠️ On *Windows*, you may need to add the paths to the *MSVC compiler (cl.exe)* and *ninja.exe* to your system environment PATH variable.
+2. **LTXV Preprocess Masks**
+Preprocesses masks for use with the LTXVideo model's latent masking. It validates mask dimensions based on VAE downscaling, supports optional inversion, handles the first frame mask separately, combines temporal masks via max pooling, applies morphological operations to grow or shrink masks, and clamps values to ensure correct opacity. The result is a set of masks ready for latent-space masking. +3. **LTXV Q8 Lora Model Loader**
+Applying LoRA to an FP8-quantized model requires special handling to preserve output quality. It's crucial to apply LoRA weights using the correct precision, as the current LoRA implementation in ComfyUI does so in a non-optimal way. This node addresses that limitation by ensuring LoRA weights are applied properly, resulting in significantly better quality. If you're working with an FP8 LTXV model, using this node guarantees that LoRA behaves as expected and delivers the intended effect. + + +# ⭐ 14.05.2025 – LTXVideo 13B 0.9.7 Distilled Release ⭐ + +### πŸš€ What's New in LTXVideo 13B 0.9.7 Distilled +1. **LTXV 13B Distilled πŸ₯³ 0.9.7**
+ Delivers cinematic-quality videos at fraction of steps needed to run full model. Only 4 or 8 steps needed for single generation.
+ πŸ‘‰ [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-13b-0.9.7-distilled.safetensors) + +2. **LTXV 13B Distilled Quantized 0.9.7**
+ Offers reduced memory requirements and even faster inference speeds. + Ideal for consumer-grade GPUs (e.g., NVIDIA 4090, 5090).
+ ***Important:*** In order to get the best performance with the quantized version please install [q8_kernels](https://github.com/Lightricks/LTXVideo-Q8-Kernels) package and use dedicated flow below.
+ πŸ‘‰ [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-13b-0.9.7-distilled-fp8.safetensors)
+ 🧩 Example ComfyUI flow available in the [Example Workflows](#example-workflows) section. + +3. **Updated LTV 13B Quantized version**
+From now on all our 8bit quantized models are running natively in ComfyUI, still with our Q8 patcher node you will get the best inference speed.
+πŸ‘‰ [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-13b-0.9.7-dev-fp8.safetensors)
+# ⭐ 06.05.2025 – LTXVideo 13B 0.9.7 Release ⭐ + +### πŸš€ What's New in LTXVideo 13B 0.9.7 + +1. **LTXV 13B 0.9.7** + Delivers cinematic-quality videos at unprecedented speed.
+ πŸ‘‰ [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-13b-0.9.7-dev.safetensors) + +2. **LTXV 13B Quantized 0.9.7** + Offers reduced memory requirements and even faster inference speeds. + Ideal for consumer-grade GPUs (e.g., NVIDIA 4090, 5090). + Delivers outstanding quality with improved performance.
+ ***Important:*** In order to run the quantized version please install [LTXVideo-Q8-Kernels](https://github.com/Lightricks/LTXVideo-Q8-Kernels) package and use dedicated flow below. Loading the model in Comfy with LoadCheckpoint node won't work.
+ πŸ‘‰ [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-13b-0.9.7-dev-fp8.safetensors)
+ 🧩 Example ComfyUI flow available in the [Example Workflows](#example-workflows) section. + +3. **Latent Upscaling Models** + Enables inference across multiple scales by upscaling latent tensors without decoding/encoding. + Multiscale inference delivers high-quality results in a fraction of the time compared to similar models.
+ ***Important:*** Make sure you put the models below in **models/upscale_models** folder.
+ πŸ‘‰ Spatial upscaling: [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-spatial-upscaler-0.9.7.safetensors).
+ πŸ‘‰ Temporal upscaling: [Download here](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltxv-temporal-upscaler-0.9.7.safetensors).
+ 🧩 Example ComfyUI flow available in the [Example Workflows](#example-workflows) section. + + +### Technical Updates + +1. ***New simplified flows and nodes***
+1.1. Simplified image to video: [Download here](example_workflows/ltxv-13b-i2v-base.json).
+1.2. Simplified image to video with extension: [Download here](example_workflows/ltxv-13b-i2v-extend.json).
+1.3. Simplified image to video with keyframes: [Download here](example_workflows/ltxv-13b-i2v-keyframes.json).
+ +# 17.04.2025 ⭐ LTXVideo 0.9.6 Release ⭐ + +### LTXVideo 0.9.6 introduces: + +1. LTXV 0.9.6 – higher quality, faster, great for final output. Download from [here](https://huggingface.co/Lightricks/LTX-Video/resolve/main/ltxv-2b-0.9.6-dev-04-25.safetensors). +2. LTXV 0.9.6 Distilled – our fastest model yet (only 8 steps for generation), lighter, great for rapid iteration. Download from [here](https://huggingface.co/Lightricks/LTX-Video/resolve/main/ltxv-2b-0.9.6-distilled-04-25.safetensors). + +### Technical Updates + +We introduce the __STGGuiderAdvanced__ node, which applies different CFG and STG parameters at various diffusion steps. All flows have been updated to use this node and are designed to provide optimal parameters for the best quality. +See the [Example Workflows](#example-workflows) section. + +# 5.03.2025 ⭐ LTXVideo 0.9.5 Release ⭐ + +### LTXVideo 0.9.5 introduces: + +1. Improved quality with reduced artifacts. +2. Support for higher resolution and longer sequences. +3. Frame and sequence conditioning (beyond the first frame). +4. Enhanced prompt understanding. +5. Commercial license availability. + +### Technical Updates + +Since LTXVideo is now fully supported in the ComfyUI core, we have removed the custom model implementation. Instead, we provide updated workflows to showcase the new features: + +1. **Frame Conditioning** – Enables interpolation between given frames. +2. **Sequence Conditioning** – Allows motion interpolation from a given frame sequence, enabling video extension from the beginning, end, or middle of the original video. +3. **Prompt Enhancer** – A new node that helps generate prompts optimized for the best model performance. + See the [Example Workflows](#example-workflows) section for more details. + +### LTXTricks Update + +The LTXTricks code has been integrated into this repository (in the `/tricks` folder) and will be maintained here. The original [repo](https://github.com/logtd/ComfyUI-LTXTricks) is no longer maintained, but all existing workflows should continue to function as expected. + +## 22.12.2024 + +Fixed a bug which caused the model to produce artifacts on short negative prompts when using a native CLIP Loader node. + +## 19.12.2024 ⭐ Update ⭐ + +1. Improved model - removes "strobing texture" artifacts and generates better motion. Download from [here](https://huggingface.co/Lightricks/LTX-Video/resolve/main/ltx-video-2b-v0.9.1.safetensors). +2. STG support +3. Integrated image degradation system for improved motion generation. +4. Additional initial latent optional input to chain latents for high res generation. +5. Image captioning in image to video [flow](example_workflows/ltxvideo-i2v.json). \ No newline at end of file From 8c21a845e6f0d15dd17fedfb5ea21326df7b53b1 Mon Sep 17 00:00:00 2001 From: elishakay Date: Thu, 10 Jul 2025 12:11:35 +0300 Subject: [PATCH 2/7] deploy on linux notes --- README.md | 4 ++ docs/deploy-on-linux.md | 109 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 113 insertions(+) create mode 100644 docs/deploy-on-linux.md diff --git a/README.md b/README.md index ccb0f9f..1a58011 100644 --- a/README.md +++ b/README.md @@ -53,6 +53,10 @@ For portable ComfyUI installations, run 1. Download [ltx-video-2b-v0.9.1.safetensors](https://huggingface.co/Lightricks/LTX-Video/blob/main/ltx-video-2b-v0.9.1.safetensors) from Hugging Face and place it under `models/checkpoints`. 2. Install one of the t5 text encoders, for example [google_t5-v1_1-xxl_encoderonly](https://huggingface.co/mcmonkey/google_t5-v1_1-xxl_encoderonly/tree/main). You can install it using ComfyUI Model Manager. +### Cloud Deployment + +In case your machine doesn't have an NVIDIA GPU, you may want to consider deploying ComfyUI (with the LTX Video Workflows) to [Vast.ai](https://cloud.vast.ai/?ref_id=276779&creator_id=276779&name=ComfyUI%20%2B%20LTX%20Video%20Lite). Check out the docs for more info about [deploying on linux](./docs/deploy-on-linux.md). + # Workflows Note that to run the example workflows, you need to have some additional custom nodes, like [ComfyUI-VideoHelperSuite](https://github.com/kosinkadink/ComfyUI-VideoHelperSuite) and others, installed. You can do it by pressing "Install Missing Custom Nodes" button in ComfyUI Manager. diff --git a/docs/deploy-on-linux.md b/docs/deploy-on-linux.md new file mode 100644 index 0000000..71eb27e --- /dev/null +++ b/docs/deploy-on-linux.md @@ -0,0 +1,109 @@ +# πŸ“¦ ComfyUI-LTXVideo Provisioning Summary + +For easily deploy with Vast.ai, check out [this template](https://cloud.vast.ai/?ref_id=276779&creator_id=276779&name=ComfyUI%20%2B%20LTX%20Video%20Lite). + +[The template script](https://gist.githubusercontent.com/ElishaKay/f92e86c2d43be9de20088991b89b0228/raw/419f67dcb0c393232d2745f63795624f6dfb0fea/ltx-video-lite.sh) provisions a full **ComfyUI environment with LTXVideo support** by downloading required nodes, models, workflows, and inputs. + +--- + +## πŸ“¦ 1. Custom Nodes (Plugins) + +These GitHub repositories are cloned into: + +``` +${COMFYUI_DIR}/custom_nodes/ +``` + +| Node Name | Source Repo | Purpose | +|-----------|-------------|---------| +| **LTXVideo** | [`Lightricks/ComfyUI-LTXVideo`](https://github.com/Lightricks/ComfyUI-LTXVideo) | Core video generation node (I2V / T2V) | +| **KJNodes** | [`kijai/ComfyUI-KJNodes`](https://github.com/kijai/ComfyUI-KJNodes) | Utility and general-purpose nodes | +| **LogicUtils** | [`aria1th/ComfyUI-LogicUtils`](https://github.com/aria1th/ComfyUI-LogicUtils) | Logic and control flow nodes | +| **Mattabyte Registry** | [`Mattabyte/ComfyUI-LTXVideo-Registry_Mattabyte`](https://github.com/Mattabyte/ComfyUI-LTXVideo-Registry_Mattabyte) | Additional registry support | +| **VideoHelperSuite** | [`Kosinkadink/ComfyUI-VideoHelperSuite`](https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite) | Frame/video utility nodes | + +--- + +## 🧠 2. Model Files + +### βœ… Checkpoint Models + +Downloaded into: + +``` +${COMFYUI_DIR}/models/checkpoints/ +``` + +| File | Source | +|------|--------| +| `ltx-video-2b-v0.9.5.safetensors` | [HuggingFace - Lightricks](https://huggingface.co/Lightricks/LTX-Video) | +| `ltxv-13b-0.9.7-distilled-fp8.safetensors` | [HuggingFace - Lightricks](https://huggingface.co/Lightricks/LTX-Video) | + +--- + +### βœ… CLIP Models + +Downloaded into: + +``` +${COMFYUI_DIR}/models/clip/ +``` + +| File | Source | +|------|--------| +| `t5xxl_fp16.safetensors` | [HuggingFace - comfyanonymous](https://huggingface.co/comfyanonymous/flux_text_encoders) | + +--- + +## πŸ–ΌοΈ 3. Input Files + +Downloaded into: + +``` +${COMFYUI_DIR}/input/ +``` + +| File | Source | +|------|--------| +| `island.jpg` | [ComfyUI Example](https://comfyanonymous.github.io/ComfyUI_examples/ltxv/island.jpg) | + +--- + +## 🧰 4. Workflows + +Downloaded into: + +``` +${COMFYUI_DIR}/user/default/workflows/ +``` + +| Workflow | Description | +|----------|-------------| +| `ltx-video-i2v-simple.json` | Image-to-Video example workflow | +| `ltx-video-t2v-simple.json` | Text-to-Video example workflow | + +--- + +## 🚫 5. Empty/Unused Model Categories + +These categories are defined in the script but **no URLs were provided**, so they are skipped during provisioning: + +- `UNET_MODELS` +- `LORA_MODELS` +- `VAE_MODELS` +- `ESRGAN_MODELS` +- `CONTROLNET_MODELS` + +--- + +## βœ… Summary of What’s Pulled + +| Component Type | Count | Destination | +|----------------|-------|-------------| +| Custom Nodes | 5 | `custom_nodes/` | +| Checkpoint Models | 2 | `models/checkpoints/` | +| CLIP Models | 1 | `models/clip/` | +| Workflows | 2 | `user/default/workflows/` | +| Input Images | 1 | `input/` | + +--- From 20dfa5dc701a1bf56812b1e804ba55aaed018c5c Mon Sep 17 00:00:00 2001 From: elishakay Date: Thu, 10 Jul 2025 12:19:41 +0300 Subject: [PATCH 3/7] downloads disclaimer --- docs/deploy-on-linux.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/docs/deploy-on-linux.md b/docs/deploy-on-linux.md index 71eb27e..b5cff81 100644 --- a/docs/deploy-on-linux.md +++ b/docs/deploy-on-linux.md @@ -2,7 +2,9 @@ For easily deploy with Vast.ai, check out [this template](https://cloud.vast.ai/?ref_id=276779&creator_id=276779&name=ComfyUI%20%2B%20LTX%20Video%20Lite). -[The template script](https://gist.githubusercontent.com/ElishaKay/f92e86c2d43be9de20088991b89b0228/raw/419f67dcb0c393232d2745f63795624f6dfb0fea/ltx-video-lite.sh) provisions a full **ComfyUI environment with LTXVideo support** by downloading required nodes, models, workflows, and inputs. +[The template script](https://gist.githubusercontent.com/ElishaKay/f92e86c2d43be9de20088991b89b0228/raw/419f67dcb0c393232d2745f63795624f6dfb0fea/ltx-video-lite.sh) provisions a full **ComfyUI environment with LTXVideo support** by downloading required nodes, models, workflows, and inputs. + +Please note: the downloading the relevant assets (listed below) can take upwards of an hour and may prevent the ComfyUI frontend from displaying during this time. To view the status of your downloads and machine setup, visit: `/#/logs`. --- From 63359138104cd93b9925250756d9fc967a74fb44 Mon Sep 17 00:00:00 2001 From: elishakay Date: Thu, 10 Jul 2025 12:24:44 +0300 Subject: [PATCH 4/7] linux template leverages just 1 model for lite setup --- docs/deploy-on-linux.md | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/docs/deploy-on-linux.md b/docs/deploy-on-linux.md index b5cff81..d9ef13c 100644 --- a/docs/deploy-on-linux.md +++ b/docs/deploy-on-linux.md @@ -4,7 +4,7 @@ For easily deploy with Vast.ai, check out [this template](https://cloud.vast.ai/ [The template script](https://gist.githubusercontent.com/ElishaKay/f92e86c2d43be9de20088991b89b0228/raw/419f67dcb0c393232d2745f63795624f6dfb0fea/ltx-video-lite.sh) provisions a full **ComfyUI environment with LTXVideo support** by downloading required nodes, models, workflows, and inputs. -Please note: the downloading the relevant assets (listed below) can take upwards of an hour and may prevent the ComfyUI frontend from displaying during this time. To view the status of your downloads and machine setup, visit: `/#/logs`. +Please note: the downloading stage of the relevant assets (listed below) can take upwards of an hour and may prevent the ComfyUI frontend from displaying during this time. To view the status of your downloads and machine setup, visit: `/#/logs`. --- @@ -39,7 +39,6 @@ ${COMFYUI_DIR}/models/checkpoints/ | File | Source | |------|--------| | `ltx-video-2b-v0.9.5.safetensors` | [HuggingFace - Lightricks](https://huggingface.co/Lightricks/LTX-Video) | -| `ltxv-13b-0.9.7-distilled-fp8.safetensors` | [HuggingFace - Lightricks](https://huggingface.co/Lightricks/LTX-Video) | --- @@ -103,7 +102,7 @@ These categories are defined in the script but **no URLs were provided**, so the | Component Type | Count | Destination | |----------------|-------|-------------| | Custom Nodes | 5 | `custom_nodes/` | -| Checkpoint Models | 2 | `models/checkpoints/` | +| Checkpoint Models | 1 | `models/checkpoints/` | | CLIP Models | 1 | `models/clip/` | | Workflows | 2 | `user/default/workflows/` | | Input Images | 1 | `input/` | From bf22746fb73a3e3bc34901e2fbf2322c1bc8dc9b Mon Sep 17 00:00:00 2001 From: elishakay Date: Thu, 10 Jul 2025 12:32:06 +0300 Subject: [PATCH 5/7] cleanup --- docs/deploy-on-linux.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/deploy-on-linux.md b/docs/deploy-on-linux.md index d9ef13c..788c1da 100644 --- a/docs/deploy-on-linux.md +++ b/docs/deploy-on-linux.md @@ -1,6 +1,6 @@ -# πŸ“¦ ComfyUI-LTXVideo Provisioning Summary +# πŸ“¦ Deploy ComfyUI-LTXVideo Flows on Linux -For easily deploy with Vast.ai, check out [this template](https://cloud.vast.ai/?ref_id=276779&creator_id=276779&name=ComfyUI%20%2B%20LTX%20Video%20Lite). +To easily deploy the ComfyUI-LTXVideo Flows with Vast.ai, check out [this template](https://cloud.vast.ai/?ref_id=276779&creator_id=276779&name=ComfyUI%20%2B%20LTX%20Video%20Lite). [The template script](https://gist.githubusercontent.com/ElishaKay/f92e86c2d43be9de20088991b89b0228/raw/419f67dcb0c393232d2745f63795624f6dfb0fea/ltx-video-lite.sh) provisions a full **ComfyUI environment with LTXVideo support** by downloading required nodes, models, workflows, and inputs. From 24fd4cd1c61ba4b06e7a5d93829f276fa7912312 Mon Sep 17 00:00:00 2001 From: elishakay Date: Thu, 10 Jul 2025 13:04:43 +0300 Subject: [PATCH 6/7] added ltxv-13b model checkpoint into linux deployment script (required by sample workflow) --- docs/deploy-on-linux.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/docs/deploy-on-linux.md b/docs/deploy-on-linux.md index 788c1da..c37b4fa 100644 --- a/docs/deploy-on-linux.md +++ b/docs/deploy-on-linux.md @@ -39,6 +39,7 @@ ${COMFYUI_DIR}/models/checkpoints/ | File | Source | |------|--------| | `ltx-video-2b-v0.9.5.safetensors` | [HuggingFace - Lightricks](https://huggingface.co/Lightricks/LTX-Video) | +| `ltxv-13b-0.9.7-distilled-fp8.safetensors` | [HuggingFace - Lightricks](https://huggingface.co/Lightricks/LTX-Video) | --- @@ -102,7 +103,7 @@ These categories are defined in the script but **no URLs were provided**, so the | Component Type | Count | Destination | |----------------|-------|-------------| | Custom Nodes | 5 | `custom_nodes/` | -| Checkpoint Models | 1 | `models/checkpoints/` | +| Checkpoint Models | 2 | `models/checkpoints/` | | CLIP Models | 1 | `models/clip/` | | Workflows | 2 | `user/default/workflows/` | | Input Images | 1 | `input/` | From 7bd9744641866b9547dc02bfcf16096793a16582 Mon Sep 17 00:00:00 2001 From: elishakay Date: Thu, 10 Jul 2025 13:36:58 +0300 Subject: [PATCH 7/7] github gist link for deployment script --- docs/deploy-on-linux.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/deploy-on-linux.md b/docs/deploy-on-linux.md index c37b4fa..8a3eba8 100644 --- a/docs/deploy-on-linux.md +++ b/docs/deploy-on-linux.md @@ -2,7 +2,7 @@ To easily deploy the ComfyUI-LTXVideo Flows with Vast.ai, check out [this template](https://cloud.vast.ai/?ref_id=276779&creator_id=276779&name=ComfyUI%20%2B%20LTX%20Video%20Lite). -[The template script](https://gist.githubusercontent.com/ElishaKay/f92e86c2d43be9de20088991b89b0228/raw/419f67dcb0c393232d2745f63795624f6dfb0fea/ltx-video-lite.sh) provisions a full **ComfyUI environment with LTXVideo support** by downloading required nodes, models, workflows, and inputs. +[The template script](https://gist.github.com/ElishaKay/f92e86c2d43be9de20088991b89b0228) provisions a full **ComfyUI environment with LTXVideo support** by downloading required nodes, models, workflows, and inputs. Please note: the downloading stage of the relevant assets (listed below) can take upwards of an hour and may prevent the ComfyUI frontend from displaying during this time. To view the status of your downloads and machine setup, visit: `/#/logs`.