
15s, 18s or 57s - The ComfyUI Edit Speed Ladder
The same edit three ways with a 4-step Lightning LoRA, Flux Kontext and full Qwen. What each trade-off actually costs.

The same edit three ways with a 4-step Lightning LoRA, Flux Kontext and full Qwen. What each trade-off actually costs.

One node takes three reference images. Person plus product, person plus scene, and the group shot - with measured timings.

One generation, seventeen outputs: the flat image, every object on transparency, and a free background plate.

We ran the same edit with and without a mask. The no-mask version won. Here’s when a mask helps and when it costs you.

Add canvas and let the model fill it. The node, the feathering, and the mistake that invents furniture.

The nine nodes behind every local instruction edit, what each does, and the one node that turns generating into editing.

Twelve generators, six prompts, one shot each, no cherry-picking, with the numbers nobody publishes: seconds per image, VRAM required, and the exact cloud price. Six of them run free on your own card.

Duix (formerly HeyGem) is the open-source HeyGen clone, it clones a face from a short video and re-lip-syncs it to any audio, offline on an NVIDIA card. We installed it and ran it. Here’s the honest result.

Eros is a baked LTX-2.3 variant that generates synced audio and video from a prompt in one model. It’s wild, but it writes its own soundtrack, so it can’t dub your voice. Here’s the honest picture.

InfiniteTalk promises infinite-length talking video from a portrait and audio, running in ComfyUI. It looks very good, but on a 32GB card the ‘infinite’ part isn’t practical yet. Here’s the honest state.