🚀 F2K9B | Flux 2 [klein] 9B LoRA Tester (FP8 Edition)

⚡ F2K9B | Flux 2 [klein] 9B LoRA Tester – v2 Update

Now with *fixed aspect ratios, modular font control, and instant strength visibility** – because details matter.*

🔥 WHAT’S NEW IN V2?

✅ Fixed Aspect Ratio Bug – No more swapped width/height! Now works perfectly for 1:1, 9:16, 16:9, etc.

✅ Modular Font Control – Set *font path** and font color centrally via Set/Get nodes.* No more hardcoded styles.

✅ LoRA Strength Burn-In – Strength now stamped below the image (e.g., Strength: 0.8). No more guessing.

🎯 KEY IMPROVEMENTS (v2 vs v1)

| Aspect Ratio | Fixed 1:1 only | Works for any ratio (9:16, 16:9, etc.) |
| Font Customization | Hardcoded | Modular via Set/Get (path + color) |
| Strength Visibility | Only in filename | Burned into image (e.g., Strength: 0.8) |
| Test Character | Lexa0Rae | Zoya0Park (new default) |

🎛️ NEW CONTROL OPTIONS (Top Left)
- Font Path – Set your preferred .ttf file via Set/Get node.
- Font Color – Choose any RGB color (default: white with black outline).
- Strength Burn-In – Now visible below the image (e.g., Strength: 0.8).

📸 EXAMPLE OUTPUT (Zoya0Park Test)

Each image now shows *Project, Epoch, and Strength** – plus custom font styling.*

🚀 HOW TO UPGRADE

1. Drag the new workflow image into ComfyUI (replaces v1).
2. Set your font path in the new Set/Get "Font Path" node.
3. Adjust font color in the Set/Get "Font Color" node.
4. Queue prompt – enjoy fixed aspect ratios and strength visibility!

🔄 Changelog (v2 vs v1)

- Fixed: Aspect ratio bug (width/height swapped).
- Added: Modular font control (path + color via Set/Get).
- Added: LoRA strength burn-in below the image.
- Updated: Default test character to Zoya0Park.

Ready to test your LoRAs with more precision? Drag the new workflow into ComfyUI and start comparing! 🚀

Professional LoRA comparison for *Flux 2 [klein] 9B (Base FP8)** – because your training deserves clarity, not guesswork.*

Optimized for:

✔ Flux2-Klein-9B-True (wikeeyang/Civitai) – Handles 4–25 steps like a champ.

✔ FP8 Precision – Fits comfortably in 16GB VRAM (even with multiple LoRAs).
✔ Lexa0Rae & Co. – Perfect for character LoRAs with consistent identity testing.

🔥 WHAT’S NEW? (vs. Flux 1 Dev)

The Flux1Dev GGUF Version: https://civitai.com/models/2385714/

✅ Centralized Sampler Hub – Set *Sampler, CFG, and Steps** in one place.* No more scattered nodes.
✅ Smart Switchboard – Toggle entire strength columns or individual LoRAs on/off. Test only what matters.
✅ Automatic Metadata Burn-In – Epoch name + LoRA strength stamped directly onto the image. No more file archaeology.
✅ Flux 2 [klein] 9B Optimized – Built for *wikeeyang’s FP8 model** (4–25 steps, no GGUF needed).*
✔ Clean Architecture – Wireless data flow via Set/Get nodes. No cable spaghetti.

🖼️ WHAT THIS WORKFLOW DOES

Turns your Flux 2 [klein] 9B LoRA training into a structured comparison matrix – with visual labels, smart saving, and flexible testing.

🎯 Key Features

| Visual Overlay | Project name, Epoch, and LoRA strength burned into the image (e.g., Lexa0Rae_Epoch12_0.8). |

| Smart Filenaming | Saves as Project_Epoch_Strength.png (e.g., Lexa0Rae_Epoch12_0.8.png). |

| Flexible Testing | Compare 5 LoRA versions × 2 strengths in one run. |

| Bypass-Friendly | Disable entire columns or individual LoRAs with a single click. |

| Step Flexibility | Works with 4–25 steps (optimized for wikeeyang’s Flux2-Klein-9B-True). |

🎛️ THE "CONTROL ROOM" (Top Left)

Everything is controlled from here – no digging through nodes.

📌 1. Model & Sampler Settings

- Checkpoint: Load Flux2-Klein-9B-True (FP8) (wikeeyang/Civitai).
- Sampler Name: Choose your preferred sampler (e.g., euler).
- CFG & Steps: Set once, applies to all columns (4–25 steps supported).

📌 2. LoRA Assets

- Load LoRAs: Select up to 5 versions (e.g., Epoch 6, 7, 8, 9, 10).
- Strength Sliders: Adjust LoRA influence (e.g., 0.6 vs 1.0).

📌 3. Text & Labels

- Project Name: Define your filename prefix (e.g., Lexa0Rae).
- Epoch Strings: Label your training steps (e.g., Epoch_6, Epoch_8).

📌 4. Switchboard (New!)

- Toggle Columns: Enable/disable entire strength columns.
- Toggle Individual LoRAs: Test only specific versions (e.g., skip Epoch 10).

⚙️ HOW TO USE

1. Download & Drag – Drag the header image (PNG) into ComfyUI. The workflow is embedded.

2. Install Missing Nodes – Use ComfyUI Manager to install:
- ComfyUI-KJNodes (Kijai) – For wireless Set/Get logic.
- ComfyUI Book Tools – For smart text overlays.
- WAS Node Suite – For image saving.
- ComfyUI-Easy-Use – For LoRA loading.
- RgThree – For optimization & bypass.
3. Set Up Control Room – Load Flux2-Klein-9B-True (FP8), your LoRAs, and labels.
4. Adjust Switchboard – Enable/disable columns as needed.
5. Queue Prompt – Watch the grid fill with labeled, organized test shots.

💡 PRO TIPS (Flux 2 [klein] 9B Edition)

- Step Range: Works best with 8–16 steps for Lexa0Rae-style character LoRAs.
- Prompting: Keep prompts structured and front-loaded (subject → action → lighting → style). Example:

Lexa0Rae, a young woman with long messy blonde hair and green eyes, wearing a black sports bra, standing in a gym, soft diffused lighting, hyperrealistic skin texture, 8k

- VRAM Efficiency: FP8 precision means you can test multiple LoRAs without OOM (even on 12GB cards).
- Consistent Seeds: Use the same seed across all columns for fair comparisons.
- Bypass Early: Disable columns you’re not testing to save time.

📸 EXAMPLE OUTPUT (Lexa0Rae Test)

Each image is labeled with *Project, Epoch, and Strength** – no more guessing!*

🚨 TROUBLESHOOTING

- Red Nodes? → Update all custom nodes via ComfyUI Manager.
- Missing LoRAs? → Double-check file paths in the LoRA loader nodes.
- Slow Performance? → Reduce batch size or disable unused columns.
- Step Issues? → Stick to 4–25 steps (optimized for wikeeyang’s model).

🔗 REQUIRED CUSTOM NODES

Install via ComfyUI Manager > "Install Missing Custom Nodes":

1. ComfyUI-KJNodes (Kijai) – For wireless Set/Get logic.
2. ComfyUI Book Tools – For smart text overlays.
3. WAS Node Suite – For image saving.
4. ComfyUI-Easy-Use – For LoRA loading.
5. RgThree – For optimization & bypass.

🎯 WHY THIS WORKFLOW?

- For Flux 2 [klein] 9B Users
- Built for wikeeyang’s FP8 model (4–25 steps, no GGUF needed).
- For LoRA Trainers – Compare epochs like a pro, not a guesser.
- For Character Artists – Test Lexa0Rae & Co. with consistent identity.
- For VRAM-Conscious Users – Efficient even with multiple LoRAs.
- For Visual Thinkers – See differences at a glance, not in filenames.

📢 Ready to test your LoRAs like a pro?

Drag the workflow image into ComfyUI and start comparing! 🚀

🔄 Changelog (vs. Flux 1 Dev Version)

The Flux1Dev GGUF Version: https://civitai.com/models/2385714/

- Added: Centralized sampler control (Sampler, CFG, Steps).
- Added: Switchboard for toggling columns/individual LoRAs.
- Added: Automatic metadata stamping (Epoch + Strength).
- Improved: Cleaner architecture with wireless Set/Get nodes.
- Optimized: For Flux2-Klein-9B-True (FP8) (4–25 steps, no GGUF).

🔥 Bonus: Lexa0Rae Prompt Template

For consistent testing, use this structure:

Lexa0Rae, a young woman with [hair description], [eye color], [outfit], [pose/action], [lighting], [background], [style], [technical specs]

Example:

Lexa0Rae, a young woman with long messy blonde hair showing dark roots, mesmerizing green eyes, wearing a black sports bra, standing in a gym with arms crossed, soft diffused lighting, hyperrealistic skin texture with visible pores, 8k