Ideogram 4: One-Model, 2 Step workflow

IG4 SOLO: Faster Ideogram 4 Generation (ONE MODEL)

YOU NEED ONLY ONE MODEL - ideogram4_fp8_scaled

v2 - with Turbo Time LoRA , 2 Steps with 4 and 8 Steps option

This workflow uses the IG4Solo node to significantly speed up your Ideogram 4 generations.

• Fast Mode: We’re running a single-pass approach - way quicker than the standard setup. Use TURBO mode (v2 works with DEFAULT and QUALITY)

• DO NOT USE ideogram4_unconditional

• The Catch: It’s a speed-demon, but it’s not a text-layer king. If you’re trying to render a complex scene with a ton of overlapping text layers, it might get a little confused. For simple, punchy, fast gens? This is your go-to.

🛠️ Installation

Option 1: ComfyUI Manager (Recommended)

1. Open ComfyUI Manager.

2. Click Install via Git URL.

3. Paste: OK and restart ComfyUI.

Option 2: Manual (CLI)

1. Navigate to your custom_nodes folder: cd path/to/ComfyUI/custom_nodes

2. Run: git clone ComfyUI.

⚠️Requirements (Don't skip this!)

• Ideogram 4 Environment: You need your Ideogram 4 model, VAE, and CLIP files ready. If you already have Ideogram 4 running in ComfyUI, you’re good to go.

• Ideogram 4 Prompt Builder (KJ)

• Updated ComfyUI: Make sure your environment is up to date.