aa

2026-06-18 00:06:15 +02:00
parent 2d0322465d
commit 1dead1c666
32 changed files with 3332 additions and 65 deletions
--- a/a6000-comfy/bootstrap.sh
+++ b/a6000-comfy/bootstrap.sh
@@ -0,0 +1,72 @@
+#!/bin/bash
+# One-time (idempotent) host setup for the Qwen-Image-Edit service on the CUDA box.
+# Runs as the service user (NO sudo). Safe to re-run: existing pieces are skipped.
+#
+# Builds, under $BASE (default ~/comfyui, outside the git repo):
+#   venv/        CUDA torch + ComfyUI deps   (RTX A6000 / sm_86)
+#   ComfyUI/     latest master + ComfyUI-GGUF custom node
+#   ComfyUI/models/{unet,text_encoders,vae}/  the v23 Q8 GGUF + encoder + vae
+#
+# CUDA lifts every MI50 workaround: no rocm wheel, no v0.3.77 pin (that pin only
+# existed because rocm-torch 2.3.1 lacks torch.library.custom_op), no lowvram.
+set -e
+
+source "$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/env.sh"
+GGUF_NODE="$COMFY/custom_nodes/ComfyUI-GGUF"
+
+echo "[bootstrap] BASE=$BASE  VENV=$VENV"
+mkdir -p "$BASE"
+
+# --- ComfyUI (latest) -------------------------------------------------------
+if [ ! -d "$COMFY/.git" ]; then
+  echo "[bootstrap] cloning ComfyUI (latest master) ..."
+  git clone --depth 1 https://github.com/comfyanonymous/ComfyUI.git "$COMFY"
+fi
+
+# --- venv + python deps -----------------------------------------------------
+# Build from the real system interpreter, NOT whatever venv is currently active.
+if [ ! -d "$VENV" ]; then
+  echo "[bootstrap] creating venv at $VENV ..."
+  /usr/bin/python3.13 -m venv "$VENV"
+fi
+source "$VENV/bin/activate"
+python -m pip install --upgrade pip wheel
+echo "[bootstrap] installing torch (CUDA cu124) ..."
+# torchaudio MUST come from the cu124 index too. Latest ComfyUI imports torchaudio
+# at startup (comfy/sd.py -> audio_vae); the PyPI default build links libcudart.so.13
+# (CUDA 13) and crashes against our cu124 torch. The matching +cu124 build fixes it.
+pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu124
+echo "[bootstrap] installing ComfyUI requirements ..."
+pip install -r "$COMFY/requirements.txt"
+# ComfyUI's requirements can re-pull a mismatched torchaudio from PyPI; re-pin it.
+pip install torchaudio --index-url https://download.pytorch.org/whl/cu124
+
+# --- ComfyUI-GGUF custom node ----------------------------------------------
+if [ ! -d "$GGUF_NODE" ]; then
+  echo "[bootstrap] cloning ComfyUI-GGUF ..."
+  git clone --depth 1 https://github.com/city96/ComfyUI-GGUF.git "$GGUF_NODE"
+fi
+pip install -r "$GGUF_NODE/requirements.txt" || pip install gguf
+
+# --- API deps ---------------------------------------------------------------
+pip install fastapi "uvicorn[standard]" websocket-client python-multipart pillow requests
+
+# --- models (resume-safe; skipped if already complete) ----------------------
+# Identical files/URLs to tour-comfy/bootstrap.sh.
+M="$COMFY/models"
+mkdir -p "$M/unet" "$M/text_encoders" "$M/vae"
+dl() {  # url  dest
+  if [ -s "$2" ]; then echo "[bootstrap] have $(basename "$2")"; return; fi
+  echo "[bootstrap] downloading $(basename "$2") ..."
+  wget -c -q --show-progress -O "$2" "$1"
+}
+dl "https://huggingface.co/Novice25/Qwen-Image-Edit-Rapid-AIO-GGUF/resolve/main/v23/Qwen-Rapid-NSFW-v23_Q8_0.gguf" \
+   "$M/unet/Qwen-Rapid-NSFW-v23_Q8_0.gguf"
+dl "https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors" \
+   "$M/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors"
+dl "https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/vae/qwen_image_vae.safetensors" \
+   "$M/vae/qwen_image_vae.safetensors"
+
+echo "[bootstrap] verifying torch + GPU ..."
+python -c "import torch; print('torch', torch.__version__, 'cuda', torch.cuda.is_available(), torch.cuda.get_device_name(0) if torch.cuda.is_available() else 'NO GPU')"
+echo "[bootstrap] BOOTSTRAP_DONE"
--- a/a6000-comfy/deploy.sh
+++ b/a6000-comfy/deploy.sh
@@ -0,0 +1,45 @@
+#!/bin/bash
+# Install/refresh systemd services for Qwen-Image-Edit on the A6000 (CUDA) box.
+# Installs 3 services + 1 target; run with sudo.
+set -e
+
+if [[ $EUID -ne 0 ]]; then
+    echo "This script must be run as root (use sudo)"
+    exit 1
+fi
+
+A6K="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"   # a6000-comfy/
+REPO="$( cd "$A6K/.." && pwd )"                            # repo root
+TOUR="$REPO/tour-comfy"
+TEMPLATES="$A6K/systemd"
+
+SVC_USER="${SUDO_USER:-$(stat -c '%U' "$A6K")}"
+SVC_GROUP="$(id -gn "$SVC_USER")"
+
+echo "Installing services:  user=$SVC_USER  group=$SVC_GROUP"
+echo "  a6k=$A6K"
+echo "  tour=$TOUR"
+
+for unit in comfyui-backend comfyui-api comfyui-watcher; do
+    sed -e "s|__USER__|$SVC_USER|g" \
+        -e "s|__GROUP__|$SVC_GROUP|g" \
+        -e "s|__A6K__|$A6K|g" \
+        -e "s|__TOUR__|$TOUR|g" \
+        "$TEMPLATES/$unit.service" > "/etc/systemd/system/$unit.service"
+    echo "  wrote /etc/systemd/system/$unit.service"
+done
+
+cp "$TEMPLATES/comfyui.target" /etc/systemd/system/comfyui.target
+echo "  wrote /etc/systemd/system/comfyui.target"
+
+echo "Reloading systemd daemon..."
+systemctl daemon-reload
+
+echo "Enabling services + target..."
+systemctl enable comfyui-backend.service comfyui-api.service comfyui-watcher.service comfyui.target
+
+echo "Starting system..."
+systemctl start comfyui.target
+
+echo "Deployment complete."
+echo "Check status with: systemctl status comfyui.target comfyui-backend comfyui-api comfyui-watcher"
--- a/a6000-comfy/env.sh
+++ b/a6000-comfy/env.sh
@@ -0,0 +1,17 @@
+#!/bin/bash
+# Shared path resolver for the Qwen-Image-Edit service on the CUDA box (RTX A6000).
+# Sourced by bootstrap.sh / run_comfyui.sh / start_api.sh.
+#
+# Unlike tour-comfy/env.sh (which derives BASE from the script location and dodges
+# an NTFS venv), this box installs into a dedicated dir OUTSIDE the git repo so the
+# ~31GB model tree + ComfyUI checkout never land in `git status`. Override with
+# COMFY_BASE / COMFY_VENV if you want it elsewhere.
+
+ENV_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"   # repo/a6000-comfy
+BASE="${COMFY_BASE:-$HOME/comfyui}"                           # install root (ext4)
+COMFY="$BASE/ComfyUI"
+VENV="${COMFY_VENV:-$BASE/venv}"
+
+# The FastAPI service + workflow are backend-agnostic; reuse them from tour-comfy
+# (pure HTTP to ComfyUI on :8188 — nothing ROCm-specific in there).
+API_DIR="$( cd "$ENV_DIR/../tour-comfy" && pwd )"
--- a/a6000-comfy/run_comfyui.sh
+++ b/a6000-comfy/run_comfyui.sh
@@ -0,0 +1,16 @@
+#!/bin/bash
+# Launch the ComfyUI backend (headless) for the Qwen-Image-Edit API on the A6000.
+# CUDA / Ampere: none of the MI50 OOM workarounds are needed. The 46GB card holds
+# the 20B Q8 model + encoder + reference latents resident, so we skip --lowvram and
+# --use-split-cross-attention (both only existed to fit 32GB) -> faster.
+set -e
+source "$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/env.sh"
+cd "$COMFY"
+source "$VENV/bin/activate"
+
+# --highvram keeps weights pinned on-GPU between requests (we have the headroom).
+exec python main.py \
+    --listen 127.0.0.1 \
+    --port 8188 \
+    --highvram \
+    "$@"
--- a/a6000-comfy/start_api.sh
+++ b/a6000-comfy/start_api.sh
@@ -0,0 +1,16 @@
+#!/bin/bash
+# Launch the FastAPI edit service (talks to the local ComfyUI on :8188).
+# Reuses tour-comfy/edit_api.py + workflow verbatim (they're backend-agnostic).
+set -e
+source "$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )/env.sh"
+source "$VENV/bin/activate"
+cd "$API_DIR"
+
+export COMFY_URL="http://127.0.0.1:8188"
+export HOST="0.0.0.0"
+export PORT="8500"
+# The A6000 is fast and not VRAM-bound on this model, so default to a full ~1MP
+# output budget (tour caps at 0.65MP to survive the MI50). Override via MAX_AREA.
+export MAX_AREA="${MAX_AREA:-1048576}"
+
+exec python edit_api.py
--- a/a6000-comfy/systemd/comfyui-api.service
+++ b/a6000-comfy/systemd/comfyui-api.service
@@ -0,0 +1,17 @@
+[Unit]
+Description=Qwen-Image-Edit FastAPI (A6000)
+After=comfyui-backend.service
+Requires=comfyui-backend.service
+
+[Service]
+Type=simple
+User=__USER__
+Group=__GROUP__
+ExecStart=/bin/bash __A6K__/start_api.sh
+Restart=on-failure
+RestartSec=5
+StandardOutput=journal
+StandardError=journal
+
+[Install]
+WantedBy=comfyui.target
--- a/a6000-comfy/systemd/comfyui-backend.service
+++ b/a6000-comfy/systemd/comfyui-backend.service
@@ -0,0 +1,16 @@
+[Unit]
+Description=ComfyUI Backend (A6000 / CUDA)
+After=network.target
+
+[Service]
+Type=simple
+User=__USER__
+Group=__GROUP__
+ExecStart=/bin/bash __A6K__/run_comfyui.sh
+Restart=on-failure
+RestartSec=5
+StandardOutput=journal
+StandardError=journal
+
+[Install]
+WantedBy=comfyui.target
--- a/a6000-comfy/systemd/comfyui-watcher.service
+++ b/a6000-comfy/systemd/comfyui-watcher.service
@@ -0,0 +1,17 @@
+[Unit]
+Description=Qwen-Image-Edit Folder Watcher (A6000)
+After=comfyui-api.service
+Requires=comfyui-api.service
+
+[Service]
+Type=simple
+User=__USER__
+Group=__GROUP__
+ExecStart=/bin/bash __TOUR__/start_watcher.sh
+Restart=on-failure
+RestartSec=5
+StandardOutput=journal
+StandardError=journal
+
+[Install]
+WantedBy=comfyui.target
--- a/a6000-comfy/systemd/comfyui.target
+++ b/a6000-comfy/systemd/comfyui.target
@@ -0,0 +1,7 @@
+[Unit]
+Description=Qwen-Image-Edit Complete System (A6000)
+Wants=comfyui-backend.service comfyui-api.service comfyui-watcher.service
+After=comfyui-watcher.service
+
+[Install]
+WantedBy=multi-user.target