Beeble SwitchX
AI generationBeebleSlowMask-guided relighting and element replacement, driven by a reference image's lighting and/or a prompt. Identity and performance preserved. Not colour grading, and not SDR to HDR.
- Runs
- On our servers
- Typical time
- 30-180s
- Inputs
- 3
- Outputs
- 1
Inputs
| Port | Type | ID |
|---|---|---|
| Prompt | text | prompt-in |
| Portrait | image | image-in |
| Lighting Referenceoptional | image | image-2 |
Outputs
| Port | Type | ID |
|---|---|---|
| Image | image | image-out |
Settings
| Setting | Type | Default | Options |
|---|---|---|---|
| Output Quality1080p is the full-quality pass. | select | 1080 | 1080p, 720p (, faster) |
Prompting
Lighting mood: "warm golden hour, cinematic, soft rim light from camera left"
Maximum prompt length: 2000 characters.
What this connects into (24)
Computed from the port compatibility rules, so it stays correct as nodes change.
- Flux LoRA
- Nano Banana 2
- Nano Banana Pro
- GPT Image 2
- Gemini Omni
- Grok Imagine 1.5
- Seedance 2.0
- MiniMax H3
- Kling 3.0
- Kling 3.0 Pro
- Grok Imagine 1.5 I2V
- Kling O3 I2V
- Kling O1 FLF2V
- WAN 2.7 I2V
- WAN Pro I2V
- Seedance 2.0 I2V
- Kling O3 Reference
- Kling O1 Reference
- Seedance 2.0 Reference
- Seedance 2.5
- Topaz Image
- SAM 3 Image
- Qwen Image Edit
- Inpaint (Flux Pro)
Node ID beeble-switchx. Open it in Studio with Tab, then type the name.