---
title: Flux-Kontext-Ultimate-LoRA
canonical_url: "https://www.modelscope.cn/models/strangerzone/Flux-Kontext-Ultimate-LoRA"
md_url: "https://www.modelscope.cn/models/strangerzone/Flux-Kontext-Ultimate-LoRA.md"
repository: strangerzone/Flux-Kontext-Ultimate-LoRA
last_updated: 2026-08-26
license: other
pipeline_tag: image-to-image
tasks:
  - image-to-image
base_model:
  - black-forest-labs/FLUX.1-Kontext-dev
base_model_relation: adapter
parameters: 3.8B
tensor_type:
  - F16
  - BF16
  - F32
library_name:
  - lora
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
supports_inference: img2img
downloads: 370
stars: 4
tags:
  - text-generation-inference
  - "template:diffusion-lora"
  - lora
  - text-image
  - "inference:false"
---

# Flux-Kontext-Ultimate-LoRA

> Flux-Kontext-Ultimate-LoRA - strangerzone 在 ModelScope 开源的模型。Flux-Kontext-Ultimate-LoRA

strangerzone/Flux-Kontext-Ultimate-LoRA 是 ModelScope 魔搭社区上的 3.8B 参数image-to-image模型，采用 other 许可，基于 black-forest-labs/FLUX.1-Kontext-dev 构建，并支持在线推理（img2img）。

- **Repository**: strangerzone/Flux-Kontext-Ultimate-LoRA
- **License**: other
- **Tasks**: image-to-image
- **Parameters**: 3.8B
- **Base model**: black-forest-labs/FLUX.1-Kontext-dev
- **Online inference**: img2img
- **Tags**: text-generation-inference, template:diffusion-lora, lora, text-image, inference:false
- **Downloads**: 370
- **Stars**: 4
- **Last updated**: 2026-08-26

Source: https://www.modelscope.cn/models/strangerzone/Flux-Kontext-Ultimate-LoRA

---

# **Flux-Kontext-Ultimate-LoRA**

> The **Flux-Kontext-Ultimate-LoRA** collection is an advanced suite of **LoRA adapters** built upon **black-forest-lab’s FLUX.1-Kontext-dev**, designed to enhance and transform visual content through perspective adaptation, cinematic lighting, artistic re-interpretation, and photo restoration. Each adapter focuses on a specific transformation mode, ensuring **natural realism, consistent lighting, and contextual fidelity** across diverse visual tasks. This collection represents a **comprehensive visual transformation toolkit** — ideal for creators, researchers, and visual artists seeking high-quality scene manipulation, enhancement, and stylistic generation.

## Included Adapters

### 1. Kontext-Top-Down-View

Transforms a scene into a **top-down perspective**, maintaining **accurate geometry, lighting consistency, and spatial realism**.
**Training Data:** 800 image pairs (400 start, 400 end)

### 2. Kontext-Bottom-Up-View

Generates a **bottom-up camera perspective**, preserving **depth, scale, and lighting direction**. Adjusts **sky or ground** naturally to match the new angle for authentic realism.
**Training Data:** 800 image pairs (400 start, 400 end)

### 3. Kontext-CAM-Left-View

Renders a **left-side viewpoint** with consistent **lighting, proportions, and textures**, revealing **previously unseen left-side details** of the scene.
**Training Data:** 800 image pairs (400 start, 400 end)

### 4. Kontext-CAM-Right-View

Produces a **right-side perspective**, preserving **natural lighting, accurate geometry, and realistic texture mapping**. Maintains environmental balance and tone.
**Training Data:** 800 image pairs (400 start, 400 end)

### 5. Kontext-Image-Upscale

Upscales low-resolution images to **4K quality**, improving **sharpness, clarity, and fine detail** while removing **noise, blur, and compression artifacts**.
**Training Data:** 1,000 image pairs (500 start, 500 end)

### 6. Kontext-Scene-Artistic

Reinterprets a scene with **creative and artistic framing**, combining **cinematic composition, surreal angles, and stylized lighting** while preserving coherence.
**Training Data:** 700 image pairs (350 start, 350 end)

### 7. FLUX.1-Kontext-Cinematic-Relighting

Applies **cinematic relighting and mood adaptation**, enhancing **contrast, light direction, and color tone** to achieve dramatic, film-grade results while maintaining realism and spatial accuracy.
**Training Data:** 1,200 image pairs (600 start, 600 end)

### 8. Monochrome-Pencil-i2i

Transforms images into **hand-drawn pencil-style monochrome sketches** while preserving **edges, contours, and lighting depth** for realistic sketch interpretations.
**Training Data:** 600 image pairs (300 start, 300 end)

### 9. Photo-Restore-i2i

Restores **damaged or aged photographs**, repairing **scratches, fading, and color degradation** while preserving authentic details and natural color balance.
**Training Data:** 900 image pairs (450 start, 450 end)

### 10. PhotoCleanser-i2i

Removes **unwanted objects, marks, or watermarks** while preserving **background details, lighting consistency, and texture continuity** for seamless image cleanup.
**Training Data:** 700 image pairs (350 start, 350 end)

### 11. Polaroid-Warm-i2i

Applies a **vintage polaroid-style color tone**, enriching images with **warm hues, soft contrast, and nostalgic lighting aesthetics** while maintaining clarity and detail.
**Training Data:** 500 image pairs (250 start, 250 end)

## Example Trigger Prompts

**Top-Down View**

> [photo content], recreate the scene from a top-down perspective. Maintain all visual proportions, lighting consistency, and realistic spatial relationships.

**Bottom-Up View**

> [photo content], recreate the scene from a bottom-up perspective. Preserve accurate depth, scale, and lighting direction to enhance realism.

**Left/Right View**

> [photo content], render the image from the left/right perspective, maintaining realistic lighting, proportions, and environmental continuity.

**Cinematic Relighting**

> [photo content], apply cinematic lighting with dynamic shadows and warm contrast for a film-like dramatic tone.

**Artistic Scene**

> [photo content], reinterpret the scene with creative composition, surreal lighting, or cinematic framing while keeping it realistic.

**Monochrome Pencil**

> [photo content], convert the scene into a detailed pencil sketch while maintaining lighting depth and edge definition.

**Photo Restore**

> [photo content], restore and enhance the image by repairing any damage, fading, or discoloration.

**Photo Cleanser**

> [photo content], remove unwanted objects or marks while keeping the lighting, texture, and background natural.

**Polaroid Warm**

> [photo content], apply a nostalgic polaroid tone with warm highlights and soft color transitions.

## Technical Information

* **Base Model:** FLUX.1-Kontext-dev (by Black Forest Labs)
* **Adapter Format:** `.safetensors`
* **Adapter Size:** 613 MB each
* **Optimization Goals:** Realism, lighting consistency, geometric fidelity, and creative flexibility
* **Framework Compatibility:** Compatible with LoRA-supported diffusion backends and context adapters