flux-txt2img

Featured

Build Flux txt2img workflows — Flux.1 Dev (SRPO), Flux 2 Klein 9B, Turbo LoRAs, FluxGuidance, and DualCLIPLoader patterns

AI & Automation 455 stars 76 forks Updated today MIT

Install

Quality Score: 94/100

Stars 20%

89

Recency 20%

100

Frontmatter 20%

70

Documentation 15%

100

Issue Health 10%

50

License 10%

100

Description 5%

100

Skill Content

# Flux Text-to-Image Workflows ## Overview Flux is a guidance-distilled diffusion model family from Black Forest Labs. It uses a separate `FluxGuidance` node instead of KSampler CFG (which must always be 1.0). Three variants are available locally: 1. **Flux.1 Dev SRPO** — Fine-tuned Flux.1 Dev with SRPO alignment. Uses DualCLIPLoader (T5XXL + CLIP-L). BF16 only. 2. **Flux 2 Klein 9B** — Distilled Flux 2 variant. Uses single CLIPLoader (Qwen3-8B) + `flux2-vae.safetensors`. Fast 4-step generation. 3. **Flux 2 Turbo LoRA** — Applied to Flux.1 Dev for 4-step generation. ## Models ### Flux.1 Dev SRPO | Component | Node | Model | Notes | |-----------|------|-------|-------| | **UNET** | `UNETLoader` | `flux.1-dev-SRPO-BFL-bf16.safetensors` | 22.7GB, **BF16 only** — FP8 produces broken results | | **CLIP** | `DualCLIPLoader` (type=`flux`) | `clip_name1`: `t5xxl_fp8_e4m3fn.safetensors`, `clip_name2`: `clip_l.safetensors` | T5XXL (4.7GB) + CLIP-L (235MB) | | **VAE** | `VAELoader` | `ae.safetensors` | Standard Flux VAE (320MB). Z-Image uses the same VAE *architecture* but different weights — its VAE is a separate file (`z-image-ae.safetensors`), not this one | ### Flux 2 Klein 9B | Component | Node | Model | Notes | |-----------|------|-------|-------| | **UNET** | `UNETLoader` | `bigLove_klein1.safetensors` | 17.3GB, Klein 9B variant | | **CLIP** | `CLIPLoader` (type=`flux`) | `qwen_3_8b_fp8mixed.safetensors` | Qwen3-8B in text_encoders/ (8.3GB) | | **VAE** | `VAELoader` | `fl...

Details

Author: artokun
Repository: artokun/comfyui-mcp
Created: 5 months ago
Last Updated: today
Language: TypeScript
License: MIT

Integrates with

Model Context Protocol · AI

Similar Skills

Semantically similar based on skill content — not just same category

AI & Automation Solid

flux2-lora-training

Comprehensive reference for training LoRAs on FLUX.2 Klein 9B and Qwen Image Edit 2511 models. Use this skill whenever the user asks about: training LoRAs for flux2/flux 2 klein/qwen-image-edit, before/after edit LoRAs (head swap, face swap, image editing), inpainting LoRAs, training at larger resolutions, latent space expansion, VAE fine-tuning, multi-reference training (2 input images → 1 output), dataset preparation for edit models, zero_cond_t, ai-toolkit/SimpleTuner/DiffSynth configs, BFS head swap LoRA methodology, Qwen Edit architecture, consistency mode, dual encoding, FuseAnyPart, ACE++, maximum training resolution, или любые вопросы об обучении диффузионных моделей. ВСЕГДА используй этот скилл. Do NOT use for writing FLUX.2 Klein generation/edit prompts at inference time (use flux2-klein-prompting), nor for general non-training diffusion architecture/inference/memory work (use diffusion-engineering); this skill is about LoRA/VAE training, not prompting or serving.

138 Updated yesterday

AI & Automation Listed

flux-image

Generate and edit images with Flux (Black Forest Labs) via AceDataCloud API. Use when creating images from text prompts, editing existing images with text instructions, or when high-quality image generation is needed. Supports multiple Flux models including dev, pro, Flux 2 variants, and kontext for editing.

13 Updated today

AI & Automation Listed

flux

The FLUX craft skill (Black Forest Labs) — generate and edit on-brand images with the right variant and license. Use when someone wants to generate images with FLUX/FLUX.2/Kontext, edit a generated image, keep a character or product consistent across a campaign, render legible text in images, get brand-exact colors, pick between FLUX variants, or asks if their FLUX use is commercially licensed. Uses the PIXEL framework. Reads image-prompt + brand-profile + design-and-templates first. The agent writes prompts/edit instructions and can call the API where connected; the HUMAN judges every image; WoopSocial publishes. License spine: [dev] outputs are commercial-OK but self-hosting for a commercial service needs a paid BFL tier; Apache-2.0 paths are [schnell]/[klein] 4B. Never use unpermitted likeness, clone trade dress, strip provenance, or invent stats. Distinct from image-prompt, ideogram/nano-banana/Midjourney (sibling tools), ai-image-editing (the edit router this feeds), and canva.

5 Updated 4 days ago

social-media-skills