Optimized Low-VRAM Multi-LoRA Workflow for Flux.2 Klein 9B (GGUF)

A refined and enhanced workflow by Experimental_dAIver and adapted for Flux.2 Klein

This workflow delivers the full power of Flux.2 Klein 9B in GGUF format, specially optimized for GPUs with less than 8 GB VRAM, like my RTX 4050 with only 6 GB. The integrated CacheDiT_Model_Optimizer and PatchSageAttentionKJ (both optional!) provide a noticeable speed boost with almost no quality loss. Full NdSuperLoraLoader support for easy multi-LoRA stacking with automatic trigger-word integration, the intuitive selectLatentSizePlus aspect-ratio and resolution selector, plus a complete SEEDVR2 Video Upscaler Subgraph for stunning high-resolution results complete this elegant and highly flexible setup.

Version 1.5 is my initial public release and brings significant improvements in speed, usability, multi-LoRA handling and upscaling quality — while remaining extremely VRAM-efficient (tested on RTX 4050 with only 6 GB).

Key features in v1.5:

• Full adaptation to Flux.2 Klein 9B with dedicated GGUF loader setup and the new Qwen3-8B GGUF text encoder

• NdSuperLoraLoader with full multi-LoRA support, auto-fetch trigger words and easy strength control

• selectLatentSizePlus — intuitive aspect-ratio and resolution selector with beautiful presets (including golden-ratio-friendly options) plus easy orientation swap

• Full SEEDVR2 Video Upscaler Subgraph — powerful DiT-based high-end upscaler that delivers stunning 4K+ results with intelligent resolution handling, Lab color correction and temporal settings. Works exceptionally well on still images too

• Improved workflow organization, expanded notes with recommended settings, and more robust saving options via SaveImageExtended

• CacheDiT and SageAttention integration for optimized performance on the smaller, faster Flux.2 Klein model

Required Custom Nodes

• ComfyUI-GGUF (UnetLoaderGGUF + CLIPLoaderGGUF)

• ComfyUI-CacheDiT (CacheDiT_Model_Optimizer)

• comfyui-kjnodes (PathchSageAttentionKJ and related nodes)

• ComfyUi-MzMaXaM (selectLatentSizePlus)

• save-image-extended-comfyui (SaveImageExtended)

• ComfyUI-SeedVR2_VideoUpscaler (full upscaler subgraph)

Models & Downloads (exact paths)

The following list explains the base models I am most frequently using with this workflow. The list as well explains where to put each file after you downloaded it.

1. Main Model (Flux.2 Klein 9B GGUF):

• File: flux2Klein-9B-Q4_1.gguf (or any other GGUF quant you prefer)

• Download: Community GGUF quants for Flux.2 Klein are available on Hugging Face (search “flux2 klein gguf” or check repositories by city96 and similar quantizers)

• Target folder: ComfyUI/models/diffusion_models/ (or Flux/ subfolder)

2. Text Encoder (CLIP)

• File: Qwen3-8B-Q4_K_M.gguf

• Download: https://huggingface.co/mradermacher/Qwen3-8B-Q4_K_M-GGUF/resolve/main/Qwen3-8B-Q4_K_M.gguf (or equivalent Qwen3-8B GGUF quant)

• Target folder: ComfyUI/models/clip/ or ComfyUI/models/text_encoders/

3. VAE

• File: full_encoder_small_decoder.safetensors (specific VAE for Flux.2 Klein setups)

• Download: Usually provided together with Flux.2 Klein GGUF packages or available from the ComfyUI community / workflow author collection

• Target folder: ComfyUI/models/vae/

4. Upscalers

• 4× Upscaler: 4x-UltraSharp → https://civarchive.com/models/116225/4x-ultrasharp

• 1× Skin-Contrast Upscaler: 1xSkinContrast-High-SuperUltraCompact.pth → https://huggingface.co/notkenski/upscalers/blob/main/1xSkinContrast-High-SuperUltraCompact.pth

• Target folder: ComfyUI/models/upscale_models/

5. LoRAs (for NdSuperLoraLoader – full multi-LoRA support)

• Any Flux-compatible LoRA (.safetensors)

• Target folder: ComfyUI/models/loras/

6. SEEDVR2 Models (for the high-end upscaler subgraph – optional but recommended)

• DiT Model: seedvr2_ema_3b-Q8_0.gguf

• VAE: ema_vae_fp16.safetensors

• Download from the official ComfyUI-SeedVR2_VideoUpscaler repository or Hugging Face and place in the folders required by the custom node.

Recommended Settings

Sampler: Euler

Scheduler: Simple or Beta

Steps: 4

CFG Scale: 1~1.5

Shift: 4–7

Resolution: freely selectable via the selectLatentSizePlus node (many golden-ratio-friendly presets included, swap orientation with one click)

How to use the workflow

1. Load the JSON in ComfyUI.

2. Select your desired aspect ratio and resolution in the selectLatentSizePlus node.

3. Enter your prompt in the Positive Prompt node (the NdSuperLoraLoader automatically handles trigger words from your LoRAs).

4. Load one or more Flux-compatible LoRAs into the NdSuperLoraLoader.

5. Generate. The workflow produces a high-quality base image.

6. (Optional) Run the SEEDVR2 upscaler subgraph for beautiful high-resolution results on images or video.

7. Use SaveImageExtended for full filename, metadata and folder control.

You can bypass the upscaler group completely for fast testing. The workflow also works with regular .safetensors Flux.2 Klein checkpoints if you replace the GGUF loaders with the standard nodes (higher VRAM usage).

This is a clean, fast and powerful base specifically tuned for the new Flux.2 Klein 9B model — perfect for quick iterations, multi-LoRA compositions and high-quality final outputs even on modest hardware.

Huge thanks to WikkedAI for the original foundation and to the creators of CacheDiT, SageAttention, NdSuperLoraLoader, selectLatentSizePlus and SeedVR2 for making Flux.2 Klein so accessible on low-VRAM GPUs.

If you have questions or want to share your results — I’m happy to hear from you in the comments. Enjoy the workflow!