Claude
38e80f8fb8
BEN2 becomes the new default local backend (clean cutouts, strong on hair/fur edges), with BiRefNet-HR available as a high-res/print alternate and U2Net kept as the lightweight fallback. Both are MIT-licensed and download weights from HuggingFace on first use (cached via the existing hf_cache bind mount), unlike U2Net/SAM which need an explicit download script. - config: new BG_REMOVAL_MODEL setting (default "ben2") - tools.py: remove-background-base64 now tries local backends in order (request.model override > BG_REMOVAL_MODEL > ben2/u2net), falling back to rembg's birefnet-general session as a last resort - requirements.gpu.txt / Dockerfile.gpu: add ben2 + transformers deps needed for the new backends, with a build-time smoke test for ben2 - frontend: model dropdown in the Remove Background dialog, threaded through api.js to the new request field Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Ro4PwQKvSc3CH19LSN21Ht
45 lines
1.8 KiB
Plaintext
45 lines
1.8 KiB
Plaintext
# =============================================================================
|
|
# GPU / Local Diffusion dependencies
|
|
# Install alongside requirements.txt when running with AI_PROVIDER=local_gpu
|
|
#
|
|
# Usage:
|
|
# pip install -r requirements.txt -r requirements.gpu.txt
|
|
#
|
|
# These are pre-installed in Dockerfile.gpu; optional in the standard image.
|
|
# =============================================================================
|
|
|
|
# HuggingFace Diffusers ecosystem
|
|
# Pinned <0.29.0: diffusers 0.29.0 added torch.xpu (Intel GPU) which fails on
|
|
# PyTorch 2.1.x with "AttributeError: module 'torch' has no attribute 'xpu'".
|
|
# Upgrade the base image in Dockerfile.gpu to pytorch 2.4+ before lifting this pin.
|
|
# (FLUX support requires diffusers>=0.29 + PyTorch>=2.4; SDXL/SD works fine here.)
|
|
diffusers>=0.28.0,<0.29.0
|
|
transformers>=4.36.0,<4.40.0
|
|
accelerate>=0.27.0
|
|
huggingface-hub>=0.23.0
|
|
safetensors>=0.4.0
|
|
|
|
# Required by SDXL pipelines
|
|
invisible-watermark>=0.2.0
|
|
omegaconf>=2.3.0
|
|
|
|
# Required by FLUX (T5 text encoder tokenizer)
|
|
sentencepiece>=0.2.0
|
|
|
|
# xformers — reduces attention VRAM ~20-30%, often unlocks the next model tier
|
|
# Must match your PyTorch+CUDA version; leave out if unsure.
|
|
# Install post-container-start if needed:
|
|
# pip install xformers --index-url https://download.pytorch.org/whl/cu121
|
|
# xformers
|
|
|
|
# Background removal — BEN2 (default, clean cutouts/hair) + BiRefNet-HR
|
|
# (high-res/print alternate). Both MIT-licensed. Verified against upstream
|
|
# source: neither requires torch>=2.5 despite the BiRefNet repo's own
|
|
# requirements.txt floor — that pin is for its training/eval scripts, not
|
|
# the inference path used here. Weights download from HuggingFace on first
|
|
# use (cached via the hf_cache bind mount, same as the diffusion models).
|
|
ben2 @ git+https://github.com/PramaLLC/BEN2.git
|
|
timm>=1.0.10
|
|
einops>=0.6.0
|
|
kornia>=0.7.0
|