Built for Production
Everything you need for expressive 360° panoramic creation, powered by diffusion models tuned for artistic impact.
14 Geometries. One Seamless Sphere.
Every shape Hextile can tile, in your hands. Drag to rotate, scroll to zoom, pick a shape, change render mode — this is the same viewer that lives inside the app.
Drag to orbit · Scroll to zoom · Switch shapes
Hexagonal Tile Diffusion
Our hexagonal tiling approach ensures seamless blending at tile boundaries, eliminating visible seams that plague traditional rectangular methods.
- Seams eliminated by construction
- Natural blending at all boundaries
- Works with any diffusion model
6 AI Pipelines
From Stable Diffusion to instruction-based editing and style transfer — choose the right pipeline for every creative goal.
- SD 3.5, SDXL, SD 2.1, SD 1.5 — proven diffusion
- Z-Image Turbo few-step fast iteration
- FLUX.2 Klein high-quality flow transformer
- FLUX.1 Schnell, Chroma1-HD, Qwen-Image, TeleStyle ALPHA
- Real-ESRGAN upscaling, Post Process & Format Converter ALPHA
- ALPHA engines are experimental and in active development — not production-ready; included for edge-case testing and feedback.
Seed Canvas
21 GPU-accelerated procedural seed generators for creating infinite seamless 360° backgrounds from pure mathematics.
- Noise, fractal, cellular & pattern generators
- Adobe Color palette import
- Live GPU preview
Super Resolution
Upscale your panoramas up to 16K+ resolution with AI-powered Real-ESRGAN. Maintain detail and quality at any scale.
- Up to 16384×8192 and beyond
- Real-ESRGAN integration
- Detail preservation
GPU Acceleration
Full CUDA support for NVIDIA GPUs, optimized for RTX 30/40 series cards.
- CUDA acceleration
- RTX optimized
- 8GB+ VRAM recommended
Local Processing
Everything runs on your machine. No cloud uploads, no data leaving your computer. Full privacy and security for sensitive projects.
- 100% offline capable
- No data uploads
- Full data privacy
Cardinal Prompt Interpolation
Give each cardinal direction its own prompt and watch them seamlessly blend across your 360° panorama — volcanic mountains to the north, frozen tundra to the south, in a single image. Write them yourself, or let the Intelligent Prompt Engine fill every direction for you.
- 5 interpolation modes: 2-Point Gradient, 3-Horizon, 4-Cardinal, 6-Direction, Tiles
- Tiles — pick any polyhedron (6–92 faces), let Qwen2.5-VL describe each face from your equirectangular image, and get N geometry-driven directional prompts that track the content as you pan. No typing — fully automatic.
- Smooth prompt blending across the sphere
- Works with all pipelines
- 231 built-in Worlds across 11 categories (Alien Planet, Cyberpunk City, Northern Fjord, Crystal Cavern...)
- Quality presets (sharp, HD, photo) & negative presets
Multi-Pass Rendering
Chain multiple diffusion passes in a single render. Each pass can use a different model, template, and strength — the output of one feeds directly into the next through GPU memory.
- Per-pass processor, template, strength, and prompt configs
- Resume from any completed pass on interruption
- Saveable multipass presets for repeatable workflows
- GPU-memory chaining between passes
Instruction-Based Editing
Describe what you want in plain English. Qwen-Image understands natural language instructions for both generating new panoramas and editing existing ones.
- Generate and Edit modes
- Natural language prompts
- ~13 GB VRAM (20B model, GGUF Q4)
Style Transfer
Apply any artistic style to your panoramas while preserving structure. TeleStyle uses diffusion-based transfer for natural, coherent results across the full 360° sphere.
- Diffusion-based style transfer
- Structure-preserving
- ~13 GB VRAM
Intelligent Prompt Engine
Don't fight with prompt syntax. Expand a simple idea into a detailed prompt tuned to the model you're using, or let the app analyze your scene and write the per-direction prompts for you.
- Model-aware expansion — phrasing & token budget tuned per pipeline (SDXL, FLUX, Qwen, SD3.5)
- Auto-write cardinal prompts from your global prompt — every direction, one click
- Local vision analysis (Florence-2 / Qwen-VL) reads your scene and fills directional prompts
- Optional Claude / OpenAI text expansion with your own API key
See It In Action
A full walkthrough of Hextile's interface — from first launch to finished 360° render.
Supported Models & Pipelines
Diffusion pipelines, utility tools, and style transfer — each optimized for different creative workflows
Stable Diffusion XL
1024px native • 12GB VRAMThe most versatile pipeline. Rich ecosystem of LoRAs and IP Adapter style transfer. ControlNet model packs ship for SD 1.5 + Qwen only.
Stable Diffusion 3.5
1024px native • 16GB VRAMLatest MMDiT architecture with superior prompt adherence and IP Adapter support. Medium, Large, and Large Turbo variants.
Stable Diffusion 1.5
512px native • 6GB VRAMThe largest community ecosystem. 8 ControlNet types, massive LoRA library, and low VRAM requirements.
Stable Diffusion 2.x
768px native • 8GB VRAMSolid mid-range option with good depth-of-field rendering characteristics. ControlNet model packs ship for SD 1.5 + Qwen only.
Qwen-Image ALPHA
Instruction-based • 13-48GB VRAM20B MMDiT with natural language editing and generation. GGUF Q4 quantization runs on 16GB GPUs. Two modes: Generate and Edit.
TeleStyle ALPHA
Style transfer • 12GB VRAMDiffusion-based artistic style transfer. Applies any style while preserving the structural content of your panorama.
Post Process ALPHA
Effects only • 0GB VRAMApply post-processing effects without AI inference. Sharpen, blur, color adjust, and more — runs instantly on CPU.
Format Converter ALPHA
Projection conversion • ~1GB VRAMGPU-accelerated conversion between 5 projection formats. Experimental — projection conversion is in active development; not production-tested for all format combinations.
Real-ESRGAN ALPHA
Any resolution • 4GB VRAMAI upscaler for boosting output resolution up to 16K+. Supports x2, x4, and anime-optimized models.
FLUX.1 Schnell ALPHA
4-step generation • 12GB VRAMUltra-fast flow matching from Black Forest Labs. High-quality output in just 4 inference steps. Apache 2.0 licensed.
Chroma1-HD ALPHA
High-res diffusion • 19GB VRAMFLUX-derived high-resolution diffusion tuned for vivid color and crisp detail — a middle ground between SDXL coherence and FLUX prompt adherence.
Z-Image Turbo
Few-step turbo • 8-16GB VRAMTurbo-distilled pipeline for fast iteration. Trade a touch of fidelity for dramatically lower step counts — ideal for rapid concept passes and prompt previews.
ALPHA ALPHA engines are experimental, in active development, and not production-ready — included for edge-case testing and feedback.
| Feature | SDXL | SD 3.5 | FLUX.1 | SD 1.5 | SD 2.x | Qwen | TeleStyle |
|---|---|---|---|---|---|---|---|
| LoRAs | 9 built-in | 3 built-in | 3 built-in | ||||
| Custom LoRAs | |||||||
| IP Adapter / Style Transfer | ViT-H | ViT-H | Diffusion | ||||
| ControlNet | 4 types | 8 types | 4 types | Union | |||
| Native Resolution | 1024px | 1024px | 1024px | 512px | 768px | Varies | 1024px |
| Min VRAM | 12 GB | 16 GB | 12 GB | 6 GB | 8 GB | 13 GB | 12 GB |
ControlNet Types by Pipeline
| ControlNet Type | SDXL | SD 1.5 | SD 2.x | Qwen-Image |
|---|---|---|---|---|
| Canny Edge detection | Union model — supports all conditioning types |
|||
| Depth Depth map conditioning | ||||
| SoftEdge Soft edge detection | ||||
| OpenPose Human pose detection | ||||
| Scribble Sketch input | ||||
| Lineart Line extraction | ||||
| MLSD Straight line detection | ||||
| Segmentation Semantic regions | ||||
| Normal Map Surface normals |
Built-in LoRA Styles
SDXL LoRAs (9 included)
- Watercolor Style
- Pixel Art XL
- Papercut Style
- Crayon Style
- Embroidery Style
- 3D Render Style
- Manga Line Art
- Voxel 3D Style
- LCM Accelerator
SD 3.5 LoRAs (3 included)
- Turbo Realism 2.0
- Photorealistic
- Anime Style
+ Install your own custom LoRAs
SD 1.5 LoRAs (3 included)
- Pixel Art
- Studio Ghibli
- LCM Accelerator
+ Thousands of community LoRAs compatible
Bring Your Own Models & LoRAs
The built-in pipelines and styles are just the start. Extend any supported family with your own LoRAs and base models — all local, all yours, no cloud lock-in.
Bring Your Own LoRAs
Drop any compatible .safetensors LoRA into your pipeline's folder and load it from the LoRA tab — including the thousands of community LoRAs from sites like Civitai. Stack multiple and tune each one's strength.
- Per-family folders: SD 1.5, SD 2.1, SD 3.5, SDXL
- Stack and blend multiple LoRAs
- Adjustable per-LoRA strength
Bring Your Own Models
Add your own base models by HuggingFace URL or import a local diffusers-format model from disk. 360 Hextile auto-detects the architecture and matches it to a supported family — you're never locked to the bundled checkpoints.
- Add via HuggingFace repository URL
- Import local diffusers-format models
- Architecture auto-detected per family
28 Post-Processing Effects
GPU-accelerated effects available on every pipeline — insertable at 4 stages: pre-process, post-tile, post-stitch, final
Color
- ACES Tonemap
- Color Grade (LUT)
- Channel Mixer
- HSL Adjustment
- Split Toning
Detail
- Detail Enhance
- Unsharp Mask
- CLAHE
- Edge-Preserving
- Guided Filter
- Bilateral Denoise
Film
- Film Grain
- Chromatic Grain
- Halation
- Sensor Noise
Lens
- Bloom
- Chromatic Aberration
Noise
- Reaction-Diffusion
- Blue Noise Dither
- Procedural Noise
- Noise Toolkit
Artistic
- Stylization
- Turing Texture
- Turing Dither
- Kuwahara
- Real-ESRGAN Detail
Finishing
- Denoise + Regrain
- Frequency Separation
System Requirements
Make sure your system meets these specifications
Minimum
- OS Windows 10 (64-bit)
- CPU Intel i5 / AMD Ryzen 5
- RAM 16 GB
- GPU NVIDIA GTX 1060 / RTX 2060 (6GB+ VRAM)
- Storage 50 GB SSD
Recommended
- OS Windows 11 (64-bit)
- CPU Intel i7 / AMD Ryzen 7 or better
- RAM 32 GB or more
- GPU NVIDIA RTX 3080 (10GB) or RTX 4090 / 5090 (24GB+ VRAM)
- Storage 100 GB NVMe SSD
Want updates when we ship new features?
One email per release. Unsubscribe in one click.
Ready to get started?
Join creators using 360 Hextile to build expressive 360° panoramic worlds.