flux-2-klein

Generate images from prompts via the RunComfy Model API using Flux 2 Klein variants.

12|2|Updated May 18, 2026

Search Tags:#image generation #text-to-image #prompting #multi-reference #flux 2 klein #runcomfy #creative iteration

Ready to run

npx skills add https://github.com/runcomfy-com/skills --skill flux-2-klein

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: flux-2-klein
Source: https://github.com/runcomfy-com/skills/tree/main/flux-2-klein
Command: npx skills add https://github.com/runcomfy-com/skills --skill flux-2-klein

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you generate high-quality images with Black Forest Labs’ Flux 2 Klein by providing model-specific prompting guidance and invoking the correct RunComfy endpoints for faster creative iteration.

Core Features & Use Cases

Low-latency concepting vs polish workflows: Uses Flux 2 Klein 4B for rapid iteration (around 4 steps) and Flux 2 Klein 9B for higher-fidelity refinement (around 25 steps).
Subject-first prompt patterns: Improves output consistency by using the model’s preferred declarative structure and specificity-focused guidance.
Multi-reference style guidance: Supports up to 4 reference images in a single call to maintain consistent brand styling and composition.
Smart routing guidance: Recommends when to switch to Flux 2 Pro, Seedream 5, GPT Image 2, or Nano Banana Pro based on quality, text, aspect, and resolution needs.
Hands-off execution via RunComfy CLI: Runs the correct API request and downloads generated outputs from approved RunComfy hosts.

Quick Start

Use the flux-2-klein skill to generate images from your prompt by saying: Generate an image of a vibrant hummingbird mid-flight sipping nectar from a bright pink hibiscus flower with cinematic lighting using Flux 2 Klein.

Frequently Asked Questions about flux-2-klein

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I generate images with Flux 2 Klein using text prompts?▼

Yes, you can use up to four reference images in a single Flux 2 Klein generation call. This multi-reference style guidance helps maintain consistent brand styling and composition across your text-to-image outputs.

What is the difference between Flux 2 Klein 4B and 9B for image generation?▼

Flux 2 Klein 4B is used for low-latency concepting with around 4 steps, while the 9B variant handles higher-fidelity polish workflows with around 25 steps. You choose between them based on whether you need rapid iteration or refined output.

How should I structure prompts for Flux 2 Klein to get better image results?▼

For better Flux 2 Klein results, use subject-first prompt patterns that focus on declarative structure and specificity. This model-specific prompting guidance improves output consistency and quality during text-to-image generation.

When should I switch from Flux 2 Klein to other image generation models?▼

You should switch from Flux 2 Klein when your specific quality, text rendering, aspect ratio, or resolution needs exceed its capabilities. The skill provides smart routing guidance to recommend alternatives like Flux 2 Pro or Seedream 5 based on these factors.

Can I use reference images for style guidance in Flux 2 Klein text-to-image workflows?▼