comfyui-launch-flags

ComfyUI launch/performance flags

Overview

ComfyUI's runtime behavior is controlled by CLI flags passed to main.py (e.g. python main.py --reserve-vram 2 --use-sage-attention). The three that matter most for making a graph run — rather than OOM or crawl — are the VRAM strategy, the attention backend, and the cache mode. This skill is the decision matrix for choosing them.

⚠️ Verification note (June 2026). Every flag below was checked against upstream comfy/cli_args.py. ComfyUI adds/renames flags often — when in doubt run python main.py --help in the target install and prefer that over this list. One common non-upstream flag: --enable-triton-backend is a SwarmUI backend flag, NOT a ComfyUI main.py flag — don't pass it to ComfyUI directly.

ℹ️ How to apply today. The MCP's start_comfyui currently replays the exact argv of the previous run — it does not compose fresh flags. So set these when you launch ComfyUI yourself (the python main.py … line, a run.bat/shell alias, or the SwarmUI backend args box), then start_comfyui will preserve them on restart. (Injecting flags through the tool is a tracked follow-up.)

---

Decide first: which flag do you need?

Symptom                                             ▶ Flag(s) to try
─────────────────────────────────────────────────────────────────────────────
CUDA out of memory, long video (LTX 2 / WAN)        ▶ --novram  (+ --cache-none)
OOM, still want models resident when they fit       ▶ --reserve-vram N  then --disable-smart-memory
GPU slows to a crawl, spills into "shared GPU        ▶ --reserve-vram 2..4
  memory" (Windows WDDM) mid-run
RAM blows up switching between models, or a huge     ▶ --cache-none
  text encoder (FLUX 2 / Mistral) won't unload
Plenty of VRAM (48GB+), want max throughput         ▶ --gpu-only  or  --highvram
Want faster sampling on NVIDIA                       ▶ --use-sage-attention   (see caveats)
Z-Image produces BLACK / wrong output               ▶ --use-pytorch-cross-attention (NOT sage)
Sage gives black output on some models              ▶ --use-pytorch-cross-attention (or fix dtype)