---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
base_model: black-forest-labs/FLUX.1-Kontext-dev
instance_prompt: >-
[photo content], render the image from the left-side perspective, keeping
consistent lighting, textures, and proportions. Maintain the realism of all
surrounding elements while revealing previously unseen left-side details
consistent with the object’s or scene’s structure.
license: other
license_name: flux-1-dev-non-commercial-license
license_link: LICENSE.md
language:
- en
pipeline_tag: image-to-image
library_name: diffusers
---

# **Kontext-CAM-Left-View**
The Kontext-CAM-Left-View is an experimental adapter for black-forest-lab's FLUX.1-Kontext-dev, designed to generate a left-side perspective of the scene while preserving consistent lighting, textures, and proportions. The model maintains the realism of all surrounding elements and accurately reveals previously unseen left-side details, ensuring seamless perspective alignment and environmental coherence. It was trained on 800 image pairs (400 start images and 400 end images) to deliver high-fidelity, geometry-consistent left-side viewpoint generation.
> [!note]
[photo content], render the image from the left-side perspective, keeping consistent lighting, textures, and proportions. Maintain the realism of all surrounding elements while revealing previously unseen left-side details consistent with the object’s or scene’s structure.
> You modified the prompt, altering its properties and subjective elements. Note: this is an experimental adapter and may contain artifacts.
---
## **Sample Inferences : Demo**
|
|
|
|
---
## Parameter Settings
| Setting | Value |
| ------------------------ | ------------------------ |
| Module Type | Adapter |
| Base Model | FLUX.1 Kontext Dev - fp8 |
| Trigger Words | [photo content], render the image from the left-side perspective, keeping consistent lighting, textures, and proportions. Maintain the realism of all surrounding elements while revealing previously unseen left-side details consistent with the object’s or scene’s structure. |
| Image Processing Repeats | 42 |
| Epochs | 22 |
| Save Every N Epochs | 1 |
Labeling: DeepCaption-VLA-7B(natural language & English)
Total Images Used for Training : 800 Image Pairs (400 Start, 400 End)
## Training Parameters
| Setting | Value |
| --------------------------- | --------- |
| Seed | - |
| Clip Skip | - |
| Text Encoder LR | 0.00001 |
| UNet LR | 0.00005 |
| LR Scheduler | constant |
| Optimizer | AdamW8bit |
| Network Dimension | 64 |
| Network Alpha | 32 |
| Gradient Accumulation Steps | - |
## Label Parameters
| Setting | Value |
| --------------- | ----- |
| Shuffle Caption | - |
| Keep N Tokens | - |
## Advanced Parameters
| Setting | Value |
| ------------------------- | ----- |
| Noise Offset | 0.03 |
| Multires Noise Discount | 0.1 |
| Multires Noise Iterations | 10 |
| Conv Dimension | - |
| Conv Alpha | - |
| Batch Size | - |
| Steps | 3300 & 400(warm up) |
| Sampler | euler |
---
## Trigger words
You should use `[photo content]` to trigger the image generation.
You should use `render the image from the left-side perspective` to trigger the image generation.
You should use `keeping consistent lighting` to trigger the image generation.
You should use `textures` to trigger the image generation.
You should use `and proportions. Maintain the realism of all surrounding elements while revealing previously unseen left-side details consistent with the object’s or scene’s structure.` to trigger the image generation.
## Download model
[Download](/prithivMLmods/Kontext-CAM-Left-View/tree/main) them in the Files & versions tab.