{"id":24191207,"url":"https://github.com/martin-rizzo/tinymodelsforlatentconversion","last_synced_at":"2026-02-06T08:38:19.751Z","repository":{"id":264883482,"uuid":"894569231","full_name":"martin-rizzo/TinyModelsForLatentConversion","owner":"martin-rizzo","description":"Command-line tools for building high-performance VAEs, latent space Transcoders, and more. Based on Tiny AutoEncoder models.","archived":false,"fork":false,"pushed_at":"2025-01-12T16:48:26.000Z","size":73,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-01-12T17:36:10.625Z","etag":null,"topics":["auto1111","comfyui","flux","stable-diffusion","text2image","transcoder","txt2img","vae"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/martin-rizzo.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-11-26T15:31:10.000Z","updated_at":"2025-01-12T16:48:29.000Z","dependencies_parsed_at":null,"dependency_job_id":"36e602ba-d732-44a1-b41b-e71d90274d72","html_url":"https://github.com/martin-rizzo/TinyModelsForLatentConversion","commit_stats":null,"previous_names":["martin-rizzo/tinyautoencoderconverter","martin-rizzo/tinymodelsforlatentconversion"],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/martin-rizzo%2FTinyModelsForLatentConversion","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/martin-rizzo%2FTinyModelsForLatentConversion/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/martin-rizzo%2FTinyModelsForLatentConversion/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/martin-rizzo%2FTinyModelsForLatentConversion/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/martin-rizzo","download_url":"https://codeload.github.com/martin-rizzo/TinyModelsForLatentConversion/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":233761319,"owners_count":18726117,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["auto1111","comfyui","flux","stable-diffusion","text2image","transcoder","txt2img","vae"],"created_at":"2025-01-13T15:17:38.123Z","updated_at":"2025-09-21T14:32:41.249Z","avatar_url":"https://github.com/martin-rizzo.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n\n# Tiny Models for Latent Conversion\n\n\u003cp\u003e\n\u003cimg alt=\"Platform\"  src=\"https://img.shields.io/badge/platform-Linux-blue\"\u003e\n\u003cimg alt=\"License\"   src=\"https://img.shields.io/github/license/martin-rizzo/TinyModelsForLatentConversion?color=blue\"\u003e\n\u003cimg alt=\"Version\"   src=\"https://img.shields.io/github/v/tag/martin-rizzo/TinyModelsForLatentConversion?label=version\"\u003e\n\u003cimg alt=\"Last\"      src=\"https://img.shields.io/github/last-commit/martin-rizzo/TinyModelsForLatentConversion\"\u003e\n\u003c!--\n|\n\u003ca href=\"https://civitai.com/models/420163/abominable-workflows\"\u003e\n   \u003cimg alt=\"CivitAI\"      src=\"https://img.shields.io/badge/page-CivitAI-00F\"\u003e\u003c/a\u003e\n\u003ca href=\"https://huggingface.co/martin-rizzo/AbominableWorkflows\"\u003e\n   \u003cimg alt=\"Hugging Face\" src=\"https://img.shields.io/badge/models-HuggingFace-yellow\"\u003e\u003c/a\u003e\n--\u003e\n\u003c/p\u003e\n\u003c/div\u003e\n\n\n**Tiny Models for Latent Conversion** offers a set of command-line utilities designed to generate three distinct types of models: ultra-fast ComfyUI-compatible Variational Autoencoders (VAEs); latent space Transcoders for seamless conversion between different latent spaces (e.g., SDXL to SD); and a custom `.safetensors` file specifically tailored for a project currently under development.\n\nAll generated models are based on the architecture and weights of the \"Tiny AutoEncoder\" models.\n\n\n## What are Tiny AutoEncoders?\n\nTiny AutoEncoders (TAEs) are highly optimized autoencoders that share the same latent space as Stable Diffusion and Flux VAEs.  This enables significantly faster and more resource-efficient image encoding and decoding.  These models were developed and trained by Ollin Boer Bohan, to whom I extend my sincere gratitude. You can find Ollin's original implementation and pre-trained models in the [Tiny AutoEncoder Repository](https://github.com/madebyollin/taesd).\n\n\n## Command-Line Tools\n\nThis project provides the following command-line conversion tools:\n- `build_tiny_vae.py`: Generates a comfyui-compatible VAE model from a Tiny AutoEncoder.\n- `build_tiny_transcoder.py`: Creates a transcoder enabling latent space conversion between different models (e.g., SDXL to SD).\n- `build_auxiliary_model.py`: Generates a custom `.safetensors` model for an ongoing project.\n\n\n## Installation and Usage\n\n1. **Clone the Repository:**  \n   First, you need to clone this repository to your local machine.\n   ```bash\n   git clone https://github.com/martin-rizzo/TinyModelsForLatentConversion.git\n   cd TinyModelsForLatentConversion\n   ```\n\n2. **Download Original Models:**  \n   Download the necessary TAESD models from Hugging Face to the `original_taesd_models` directory:\n   ```bash\n   ./download_original_models.sh\n   ```\n   For more information on what TAESD models are needed and how to download them manually, refer to the [original_taesd_models/README.md](original_taesd_models/README.md) documentation.\n\n3. **Create the Virtual Environment and Install Dependencies:**  \n   The `build_tiny_vae.sh` script came with a virtual environment setup argument, so you don't need to create a virtual environment manually.\n   ```bash\n   ./build_tiny_vae.sh --create-venv\n   ```\n\n4. **Run the Bash Script:**  \n   Execute the `makeall.sh` script to generate all VAEs and transcoders automatically:\n   ```bash\n   ./makeall.sh\n   ```\n\n\n## Output\n\nWhen running the general `makeall.sh` script, it will generate several files in the `output` directory, including VAE models and transcoders for different latent spaces. The generated files will have names like:\n- `tiny_vae_*.safetensors`: VAE models compatible with ComfyUI.\n- `transcoder_from_*_to_*.safetensors`: Transcoder model files.\n- `auxiliary_model.safetensors`: Custom auxiliary model for a specific project.\n\n\n## License\n\n**Copyright (c) 2024-2025 Martin Rizzo**  \nThis project is licensed under the MIT license.  \nDetails can be found in the [\"LICENSE\"](LICENSE) file.\n\n\n## Disclaimer\n\nThis tool is provided \"as is\" without any warranty. Use at your own risk.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmartin-rizzo%2Ftinymodelsforlatentconversion","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmartin-rizzo%2Ftinymodelsforlatentconversion","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmartin-rizzo%2Ftinymodelsforlatentconversion/lists"}