ComfyUI-EfficientTAM

ComfyUI-EfficientTAM requires a minimum of 128MB (0.1GB) of VRAM for base operation. For optimal performance, a GPU with at least 4GB of VRAM is recommended. This node supports low VRAM mode for resource-constrained setups.

✅ RTX 3060 (12GB): Yes, fully compatible with 10.7GB headroom. ✅ RTX 4070 (12GB): Yes, fully compatible with 10.7GB headroom. ✅ RTX 4070 Ti (16GB): Yes, fully compatible with 14.3GB headroom. ✅ RTX 4090 (24GB): Yes, fully compatible with 21.5GB headroom

ComfyUI-EfficientTAM requires the following PyTorch-related packages: torch=2.5.1+cu124=pypi_0, torchaudio=2.5.1+cu124=pypi_0, torchvision=0.20.1+cu124=pypi_0. Ensure your ComfyUI environment has these installed. CUDA 12.4 is required. This is optimized for RTX 4000 series and newer GPUs.

To run ComfyUI-EfficientTAM, you need to install: _libgcc_mutex=0.1=main, _openmp_mutex=5.1=1_gnu, aiofiles=23.2.1=pypi_0, aiohappyeyeballs=2.4.3=pypi_0, aiohttp=3.11.7=pypi_0, aiosignal=1.3.1=pypi_0, altair=5.5.0=pypi_0, annotated-types=0.7.0=pypi_0, antlr4-python3-runtime=4.9.3=pypi_0, anyio=3.7.1=pypi_0, aom=3.6.0=h6a678d5_0, attrs=24.2.0=pypi_0, bcrypt=4.2.1=pypi_0, blas=1.0=mkl, brotli-python=1.0.9=py312h6a678d5_8, build=1.2.2.post1=pypi_0, bzip2=1.0.8=h5eee18b_6, ca-certificates=2024.8.30=hbcca054_0, cairo=1.16.0=hb05425b_5, certifi=2024.8.30=pyhd8ed1ab_0, cffi=1.17.1=pypi_0, charset-normalizer=3.3.2=pyhd3eb1b0_0, click=8.1.7=pypi_0, cmake=3.31.1=pypi_0, contourpy=1.3.1=pypi_0, cryptography=43.0.3=pypi_0, cycler=0.12.1=pypi_0, dav1d=1.2.1=h5eee18b_0, decorator=5.1.1=pyhd3eb1b0_0, defusedxml=0.7.1=pypi_0, einops=0.8.0=pypi_0, expat=2.6.2=h6a678d5_0, fastapi=0.115.5=pypi_0, ffmpeg=4.2.2=h167e202_0, ffmpy=0.4.0=pypi_0, filelock=3.16.1=pypi_0, flash-attn=2.7.0.post2=pypi_0, fontconfig=2.14.1=h4c34cd2_2, fonttools=4.55.0=pypi_0, freetype=2.12.1=h4a9f257_0, frozenlist=1.5.0=pypi_0, fsspec=2024.10.0=pypi_0, giflib=5.2.1=h5eee18b_3, glib=2.78.4=h6a678d5_0, glib-tools=2.78.4=h6a678d5_0, gmp=6.2.1=h58526e2_0, gnutls=3.6.13=h85f3911_1, gradio=4.44.0=pypi_0, gradio-client=1.3.0=pypi_0, gradio-image-prompter=0.1.0=pypi_0, graphite2=1.3.14=h295c915_1, h11=0.14.0=pypi_0, harfbuzz=4.3.0=hf52aaf7_2, httpcore=1.0.7=pypi_0, httpx=0.27.0=pypi_0, huggingface-hub=0.26.2=pypi_0, hydra-core=1.3.2=pypi_0, icu=73.1=h6a678d5_0, idna=3.7=py312h06a4308_0, imageio=2.9.0=pyhd3eb1b0_0, imageio-ffmpeg=0.5.1=pyhd8ed1ab_0, importlib-resources=6.4.5=pypi_0, iniconfig=2.0.0=pypi_0, intel-openmp=2023.1.0=hdb19cb5_46306, iopath=0.1.10=pypi_0, jinja2=3.1.4=pypi_0, jpeg=9e=h5eee18b_3, jsonschema=4.23.0=pypi_0, jsonschema-specifications=2024.10.1=pypi_0, kiwisolver=1.4.7=pypi_0, lame=3.100=h7b6447c_0, lcms2=2.12=h3be6417_0, ld_impl_linux-64=2.38=h1181459_1, leptonica=1.82.0=h42c8aad_2, lerc=3.0=h295c915_0, libarchive=3.6.2=h6ac8c49_3, libdeflate=1.17=h5eee18b_1, libffi=3.4.4=h6a678d5_1, libgcc-ng=11.2.0=h1234567_1, libglib=2.78.4=hdc74915_0, libgomp=11.2.0=h1234567_1, libiconv=1.16=h5eee18b_3, libogg=1.3.5=h27cfd23_1, libopus=1.3.1=h7b6447c_0, libpng=1.6.39=h5eee18b_0, libstdcxx-ng=11.2.0=h1234567_1, libtheora=1.1.1=h7f8727e_3, libtiff=4.5.1=h6a678d5_0, libuuid=1.41.5=h5eee18b_0, libvorbis=1.3.7=h7b6447c_0, libvpx=1.13.1=h6a678d5_0, libwebp=1.3.2=h11a3e52_0, libwebp-base=1.3.2=h5eee18b_0, libxcb=1.15=h7f8727e_0, libxml2=2.10.4=hfdd30dd_2, linkify-it-py=2.0.3=pypi_0, lz4-c=1.9.4=h6a678d5_1, markdown-it-py=2.2.0=pypi_0, markupsafe=2.1.5=pypi_0, matplotlib=3.9.2=pypi_0, mdit-py-plugins=0.3.3=pypi_0, mdurl=0.1.2=pypi_0, mkl=2023.1.0=h213fc3f_46344, mkl-service=2.4.0=py312h5eee18b_1, mkl_fft=1.3.10=py312h5eee18b_0, mkl_random=1.2.7=py312h526ad5a_0, moviepy=1.0.3=pyhd8ed1ab_1, mpmath=1.3.0=pypi_0, multidict=6.1.0=pypi_0, narwhals=1.14.2=pypi_0, ncurses=6.4=h6a678d5_0, nettle=3.6=he412f7d_0, networkx=3.4.2=pypi_0, ninja=1.11.1.2=pypi_0, numpy=1.26.4=pypi_0, nvidia-cublas-cu12=12.4.5.8=pypi_0, nvidia-cuda-cupti-cu12=12.4.127=pypi_0, nvidia-cuda-nvrtc-cu12=12.4.127=pypi_0, nvidia-cuda-runtime-cu12=12.4.127=pypi_0, nvidia-cudnn-cu12=9.1.0.70=pypi_0, nvidia-cufft-cu12=11.2.1.3=pypi_0, nvidia-curand-cu12=10.3.5.147=pypi_0, nvidia-cusolver-cu12=11.6.1.9=pypi_0, nvidia-cusparse-cu12=12.3.1.170=pypi_0, nvidia-nccl-cu12=2.21.5=pypi_0, nvidia-nvjitlink-cu12=12.4.127=pypi_0, nvidia-nvtx-cu12=12.4.127=pypi_0, omegaconf=2.3.0=pypi_0, opencv-python=4.10.0.84=pypi_0, openh264=2.1.1=h4ff587b_0, openjpeg=2.5.2=he7f1fd0_0, openssl=3.0.15=h5eee18b_0, orjson=3.10.12=pypi_0, packaging=24.2=pypi_0, pandas=2.2.3=pypi_0, paramiko=3.5.0=pypi_0, pcre2=10.42=hebb0a14_1, pillow=10.4.0=py312h5eee18b_0, pip=24.0=py312h06a4308_0, pixman=0.40.0=h7f8727e_1, pluggy=1.5.0=pypi_0, portalocker=3.0.0=pypi_0, proglog=0.1.9=py_0, propcache=0.2.0=pypi_0, psutil=5.9.8=pypi_0, pycparser=2.22=pypi_0, pycryptodome=3.21.0=pypi_0, pydantic=2.10.2=pypi_0, pydantic-core=2.27.1=pypi_0, pydub=0.25.1=pypi_0, pygments=2.18.0=pypi_0, pynacl=1.5.0=pypi_0, pyparsing=3.2.0=pypi_0, pyproject-hooks=1.2.0=pypi_0, pysocks=1.7.1=py312h06a4308_0, pytest=8.3.3=pypi_0, python=3.12.4=h5148396_1, python-dateutil=2.9.0.post0=pypi_0, python-multipart=0.0.12=pypi_0, pytz=2024.2=pypi_0, pyyaml=6.0.2=pypi_0, readline=8.2=h5eee18b_0, referencing=0.35.1=pypi_0, requests=2.32.3=py312h06a4308_0, rich=13.9.4=pypi_0, rpds-py=0.21.0=pypi_0, ruff=0.8.0=pypi_0, safehttpx=0.1.1=pypi_0, scipy=1.14.1=pypi_0, semantic-version=2.10.0=pypi_0, setuptools=72.1.0=py312h06a4308_0, shellingham=1.5.4=pypi_0, six=1.16.0=pypi_0, sniffio=1.3.1=pypi_0, spaces=0.30.4=pypi_0, sqlite=3.45.3=h5eee18b_0, starlette=0.41.3=pypi_0, supervision=0.25.0=pypi_0, sympy=1.13.1=pypi_0, tbb=2021.8.0=hdb19cb5_0, tesseract=5.2.0=h6a678d5_0, tk=8.6.14=h39e8969_0, tomlkit=0.12.0=pypi_0, torch=2.5.1+cu124=pypi_0, torchaudio=2.5.1+cu124=pypi_0, torchvision=0.20.1+cu124=pypi_0, tqdm=4.67.1=pypi_0, triton=3.1.0=pypi_0, typer=0.13.1=pypi_0, typing-extensions=4.12.2=pypi_0, tzdata=2024.2=pypi_0, uc-micro-py=1.0.3=pypi_0, urllib3=2.2.2=py312h06a4308_0, uvicorn=0.32.1=pypi_0, websockets=11.0.3=pypi_0, wheel=0.43.0=py312h06a4308_0, x264=1!152.20180806=h14c3975_0, xz=5.4.6=h5eee18b_1, yarl=1.18.0=pypi_0, zlib=1.2.13=h5eee18b_1, zstd=1.5.5=hc292b87_2. You can install these using pip or add them to your requirements.txt file.

ComfyUI-EfficientTAM supports low VRAM mode. To reduce memory usage: (1) Enable --lowvram or --medvram flags in ComfyUI, (2) Reduce batch size to 1, (3) Use fp16 or fp8 precision if supported, (4) Close other GPU applications.

To install ComfyUI-EfficientTAM: (1) Navigate to your ComfyUI/custom_nodes directory, (2) Clone the repository: git clone https://github.com/ryanontheinside/ComfyUI_EfficientTAM, (3) Install dependencies: pip install -r requirements.txt (if present), (4) Restart ComfyUI. Alternatively, use ComfyUI Manager for one-click installation.