Comfyui Character Gen

If you're building character generation workflows in ComfyUI and need consistent faces across images or video, this walks you through the decision tree. It covers the major identity preservation methods (InfiniteYou, FLUX Kontext, PuLID, InstantID, IP-Adapter) and tells you which to pick based on your use case: zero-shot generation from references, 3D render conversion to photorealistic portraits, iterative editing without retraining, or video with voice. The workflows are concrete, node-level templates with CFG settings and VRAM requirements. Updated for 2026 models including FLUX.2 multi-reference support and RTX 50 series optimizations. It won't help with general image generation, model training, or debugging your existing workflows, but if you need a face to stay the same across outputs, it's a solid starting point.

skills.sh installs: 764
GitHub stars: 94
last updated: Jun 3, 2026

View on GitHub

Install

$ npx -y skills add mckruz/comfyui-expert --skill comfyui-character-gen --agent claude-code

Installs into .claude/skills of the current project.

Files

ComfyUI Character Generation Expert

Build production-ready ComfyUI workflows for consistent character generation across image, video, and voice modalities.

Quick Decision: Which Approach?

Starting from reference images (like 3D renders)? → InfiniteYou (state-of-the-art 2025) or InstantID + IP-Adapter (proven, lower VRAM)

Need highest identity fidelity? → FLUX.2 (NEW 2026: up to 10 ref images) or PuLID Flux II (no model pollution)

Want iterative editing without retraining? → FLUX Kontext (context-aware, maintains consistency across edits)

Creating video content? → LTX-2 (NEW 2026: 4K production-ready), Wan 2.2 MoE (film-level), or FramePack (60-sec on 6GB!)

Need voice for character? → TTS Audio Suite (unified platform, 23 languages) or F5-TTS Cross-Lingual (NEW 2026)

Core Workflow Patterns

Pattern 1: Zero-Shot Character Generation (No Training)

Best for: Quick iteration, 3D-to-photorealism conversion, limited reference images

Load Reference Face → InstantID + IP-Adapter FaceID → ControlNet Pose → KSampler → FaceDetailer → Upscale

SKILL.md on GitHub

Details

Repository: github.com/mckruz/comfyui-expert
Category: Generative Media
First seen: Jun 3, 2026

More from mckruz/comfyui-expert

All 10