{"id":16277508,"url":"https://github.com/pythops/jetfusion","last_synced_at":"2025-09-19T04:32:40.730Z","repository":{"id":213336715,"uuid":"732436229","full_name":"pythops/jetfusion","owner":"pythops","description":"Diffusion models on Nvidia Jetson boards","archived":false,"fork":false,"pushed_at":"2024-01-28T15:25:58.000Z","size":3,"stargazers_count":4,"open_issues_count":0,"forks_count":1,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-01-04T20:47:03.022Z","etag":null,"topics":["diffusion-models","jetson","nvidia"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/pythops.png","metadata":{"files":{"readme":"Readme.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2023-12-16T17:05:15.000Z","updated_at":"2024-06-15T20:19:41.000Z","dependencies_parsed_at":"2023-12-20T14:16:07.916Z","dependency_job_id":"1e52be77-60a1-4ad3-ab63-7171910d44ac","html_url":"https://github.com/pythops/jetfusion","commit_stats":null,"previous_names":["pythops/jetfusion"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pythops%2Fjetfusion","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pythops%2Fjetfusion/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pythops%2Fjetfusion/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pythops%2Fjetfusion/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/pythops","download_url":"https://codeload.github.com/pythops/jetfusion/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":233551898,"owners_count":18693053,"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":["diffusion-models","jetson","nvidia"],"created_at":"2024-10-10T18:55:13.506Z","updated_at":"2025-09-19T04:32:40.473Z","avatar_url":"https://github.com/pythops.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Diffusion models on Jetson boards\n\nEasy deploy of diffusion models on Nvidia jetson boards.\n\n## 🧰 Requirement\n\n- Nvidia Jetson board with at least 8GB of Memory.\n- At least 25GB free space on the disk.\n\n## 🛠️ Setup\n\n1. Download and install the minamilist image for your board [here🔗](https://github.com/pythops/jetson-image)\n\n2. Install the necessary dependencies\n\n```\nsudo apt update \u0026\u0026 \\\nsudo apt install -y \\\n    cuda-toolkit-11-4 \\\n    libcudnn8-dev \\\n    libnuma-dev \\\n    libopenblas-dev \\\n    autoconf \\\n    build-essential \\\n    g++-8 \\\n    gcc-8 \\\n    clang-8 \\\n    lld-8 \\\n    gettext-base \\\n    gfortran-8 \\\n    libbz2-dev \\\n    libc++-dev \\\n    libcgal-dev \\\n    libffi-dev \\\n    libfreetype6-dev \\\n    libhdf5-dev \\\n    libjpeg-dev \\\n    liblzma-dev \\\n    libncurses5-dev \\\n    libncursesw5-dev \\\n    libreadline-dev \\\n    libssl-dev \\\n    libxslt-dev \\\n    openssl \\\n    python-openssl \\\n    scons \\\n    python3-pip \\\n    python-is-python3 \\\n    libopenblas-dev \\\n    git-lfs \u0026\u0026 \\\npip install --user poetry\n```\n\n3. Add `$USER` (by default `jetson`) user to the video group\n\n```\nsudo usermod -aG video $USER\n```\n\n\u003e You need to re-login for the changes to take effect.\n\n4. Clone `sd-turbo` diffusion model\n\n```\nmkdir -p ~/diffusion/sd-turbo \u0026\u0026 \\\ngit lfs install --skip-repo \u0026\u0026 \\\ngit clone https://huggingface.co/stabilityai/sd-turbo ~/diffusion/sd-turbo\n\n```\n\n4. Clone the jetfusion repository\n\n```\ngit clone https://github.com/pythops/jetfusion \u0026\u0026 cd jetfusion\n```\n\n4. Install the python dependencies\n\n```\npoetry install --no-root\n```\n\n5. Run the api server\n\n```\nDIFFUSION_MODEL_DIR=\"$HOME/diffusion/sd-turbo\" poetry run uvicorn --host 0.0.0.0 api:api\n```\n\n## 🪃 Usage\n\nTo display images in your terminal, you'll need to install [img2sixel](https://github.com/saitoha/libsixel)\n\nYour terminal should support the sixel graphics format, check here for more infos https://www.arewesixelyet.com/\n\n### Using curl\n\n```\ncurl -s https://localhost:8000/generate \\\n    -H \"Content-Type: application/json\" \\\n    -d '{\"prompt\": \"Colorful fish in the coral reef\"}' | img2sixel\n```\n\n### Using httpie\n\n```\nhttp https://localhost:8000/generate \\\nprompt=\"Colorful fish in the coral reef\" | img2sixel\n```\n\n## ⚖️ License\n\nAGPLv3\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpythops%2Fjetfusion","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpythops%2Fjetfusion","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpythops%2Fjetfusion/lists"}