Skip to main content

Roadmap

Last Updated: November 18, 2025 | Timeline: November 2025 - April 2026

Roadmap Overview​

┌─────────────────────────────────────────────────────────────────────────┐
│ COMPLETED (Q4 2025 - November 2025)                                    │
├─────────────────────────────────────────────────────────────────────────┤
│ ✓ Repository rebranded: TryOnDiffusion → OpenTryOn                      │
│ ✓ Datasets module (Fashion-MNIST, VITON-HD) with class-based adapters   │
│ ✓ Virtual Try-On adapters: Amazon Nova Canvas, Kling AI, Segmind        │
│ ✓ Enhanced documentation                                                 │
└─────────────────────────────────────────────────────────────────────────┘

┌─────────────────────────────────────────────────────────────────────────┐
│ IN PROGRESS (November 17-30, 2025)                                      │
├─────────────────────────────────────────────────────────────────────────┤
│ • Dataset module testing and refinement                                 │
│ • API adapter pattern standardization                                   │
│ • Documentation improvements                                             │
│ • Image Generation model integrations                                   │
│    - Closed-source: Nano Banana, GPT-Image-1, FLUX Kontext Pro          │
│    - Open-source: FLUX.1-dev and other open-source models               │
└─────────────────────────────────────────────────────────────────────────┘

┌─────────────────────────────────────────────────────────────────────────┐
│ PLANNED (December 2025 - January 2026)                                  │
├─────────────────────────────────────────────────────────────────────────┤
│ Core API Infrastructure                                                 │
│    • Base provider classes/interfaces (BaseProvider, BaseVTONAdapter)    │
│    • Unified API response types (GenerationResponse, VTONResponse)      │
│    • Configuration management, error handling, retry logic               │
│    • VirtualTryOn unified interface                                     │
│    • Standardize existing adapters (Nova Canvas, Kling AI, Segmind)    │
│                                                                          │
│ Image Generation & Model Integration (December 2025)                   │
│    • ImageGenerator unified interface                                   │
│    • Model registry system                                              │
│    • Additional open-source image generation models                     │
│                                                                          │
│ Video Generation Capabilities                                          │
│    • VideoGenerator unified interface                                   │
│    • Closed-source providers: SORA2, Kling AI, LUMA Ray                │
│    • Open-source models from GitHub, HuggingFace                        │
│                                                                          │
│ Open-Source Virtual Try-On Models                                       │
│    • Image VTON: Leffa, DreamO, Voost, IDM-VTON, CatVTON               │
│    • Video VTON: Video VTON models                                     │
│    • 3D VTON: Open-source 3D virtual try-on models                     │
│    • Integration with HuggingFace and GitHub models                    │
│                                                                          │
│ Agentic AI System for PDP Optimization                                 │
│    • Multi-agent system for product detail page evaluation             │
│    • LLM and VLM integration for analysis and optimization             │
│    • Automated PDP content analysis and recommendations                │
│                                                                          │
│ Testing & Quality                                                       │
│    • Unit/integration tests, >85% coverage                              │
└─────────────────────────────────────────────────────────────────────────┘

┌─────────────────────────────────────────────────────────────────────────┐
│ EXPLORING (February - April 2026)                                       │
├─────────────────────────────────────────────────────────────────────────┤
│ • Additional providers: Fashn AI, TryOnLabs, Qwen-Image                │
│ • HuggingFace model loading utilities                                   │
│ • CLI improvements (`ot generate`, `ot tryon`, `ot video`, `ot agent`) │
│ • Additional datasets (DeepFashion, FashionGen)                        │
│ • Enhanced agentic AI capabilities                                      │
│ • Advanced features: async/await, caching, batch processing (if time)  │
└─────────────────────────────────────────────────────────────────────────┘

Detailed Breakdown​

Completed (Q4 2025 - November 2025)​

  • Repository rebranding, documentation updates
  • Datasets Module: Fashion-MNIST, VITON-HD with class-based adapter pattern
  • Virtual Try-On Adapters: Amazon Nova Canvas, Kling AI, Segmind

In Progress (November 17-30, 2025)​

  • Dataset module testing and refinement
  • API adapter pattern standardization
  • Documentation improvements
  • Image Generation Model Integrations:
  • Nano Banana image generation adapter (closed-source)
  • GPT-Image-1 adapter (closed-source)
  • FLUX Kontext Pro adapter (closed-source)
  • FLUX.1-dev adapter (open-source)
  • Additional open-source image generation models from HuggingFace/GitHub

Planned (December 2025 - January 2026)​

December 2025: Core API Infrastructure

  • Base provider classes/interfaces (BaseProvider, BaseVTONAdapter)
  • Unified API response types (GenerationResponse, VTONResponse)
  • Configuration management, error handling, retry logic
  • VirtualTryOn unified interface
  • Standardize existing adapters (Nova Canvas, Kling AI, Segmind)

December 2025: Image Generation Unified Interface

  • ImageGenerator unified interface
  • Model registry system for image generation models
  • Standardize image generation adapters
  • Unit/integration tests for image generation, >90% coverage

January - February 2026: Video Generation & Agentic AI

  • VideoGenerator unified interface
  • Video providers: SORA2, Kling AI, LUMA Ray (closed-source)
  • Open-source video models from GitHub, HuggingFace
  • Agentic AI system for PDP (Product Detail Page) optimization
  • LLM and VLM integration for content analysis
  • Automated PDP evaluation and optimization pipeline

January - February 2026: Open-Source Virtual Try-On Models

  • Image VTON models: Leffa, DreamO, Voost, IDM-VTON, CatVTON
  • Video VTON models integration
  • 3D VTON models integration
  • HuggingFace model loading utilities for VTON models
  • GitHub model integration framework

Exploring (February - April 2026)​

  • Additional providers: Fashn AI, TryOnLabs, Qwen-Image
  • Additional open-source VTON models from community
  • CLI improvements (ot generate, ot tryon, ot edit, ot list-models, ot video, ot agent)
  • Additional datasets (DeepFashion, FashionGen)
  • Advanced features: async/await, caching, batch processing (if time permits)
  • Enhanced agentic AI capabilities and integrations

Success Metrics​

November 2025: ✓ Datasets module (2 datasets), ✓ 3 VTON adapters (Nova Canvas, Kling AI, Segmind), Image generation adapters (Nano Banana, GPT-Image-1, FLUX Kontext Pro, FLUX.1-dev), >80% test coverage

December 2025: Base architecture, Unified VirtualTryOn interface, ImageGenerator unified interface, Model registry system, >85% test coverage

January - February 2026: VideoGenerator interface, Video providers (SORA2, Kling AI, LUMA Ray), Open-source VTON models (Leffa, DreamO, Voost, IDM-VTON, CatVTON, Video VTON, 3D VTON), Agentic AI system for PDP optimization, >90% coverage

February - April 2026: Additional providers, CLI improvements, Enhanced DX, Advanced agentic AI features

Contributor Roadmap​

For Contributors: This section highlights areas where we need your help! Pick a task that matches your skills and interests.

Good First Issues (Great for New Contributors)​

These are perfect for getting started with the project:

Documentation & Examples​

  • Documentation: Improve README examples and add code snippets
  • Tutorials: Create Jupyter notebook tutorials for datasets module
  • Examples: Add usage examples for each API adapter
  • Type Hints: Add type hints to existing functions and classes
  • Labels: good first issue, documentation, help wanted

Testing & Quality​

  • Unit Tests: Write tests for dataset loaders (Fashion-MNIST, VITON-HD)
  • Integration Tests: Add tests for API adapters (Nova Canvas, Kling AI, Segmind)
  • Test Coverage: Improve test coverage for existing modules
  • Labels: good first issue, testing, help wanted

Code Quality​

  • Linting: Fix linting errors and improve code style
  • Error Handling: Improve error messages and exception handling
  • Code Comments: Add docstrings and inline comments
  • Labels: good first issue, code quality, help wanted

Intermediate Contributions​

For contributors with some experience:

API Adapters & Integrations​

  • Base Classes: Implement BaseProvider and BaseVTONAdapter abstract classes
  • Response Types: Create unified response types (GenerationResponse, VTONResponse)
  • Error Handling: Implement retry logic with exponential backoff
  • Configuration: Build configuration management system
  • Labels: intermediate, api, help wanted

Image Generation Providers (November 2025 - High Priority)​

  • Nano Banana: Implement Nano Banana image generation adapter (closed-source)
  • GPT-Image-1: Integrate GPT-Image-1 API (closed-source)
  • FLUX Kontext Pro: Add FLUX Kontext Pro adapter (closed-source)
  • FLUX.1-dev: Create adapter for open-source FLUX.1-dev model
  • Additional Open-Source Models: Integrate other open-source image generation models from HuggingFace/GitHub
  • Labels: intermediate, image-generation, help wanted, high-priority

Dataset Adapters​

  • DeepFashion: Create DeepFashion dataset adapter
  • FashionGen: Implement FashionGen dataset loader
  • Dataset Utils: Add preprocessing utilities (masks, segmentation, pose detection)
  • Labels: intermediate, datasets, help wanted

Open-Source Virtual Try-On Models​

  • Leffa: Integrate Leffa image virtual try-on model
  • DreamO: Integrate DreamO image virtual try-on model
  • Voost: Integrate Voost image virtual try-on model
  • IDM-VTON: Integrate IDM-VTON image virtual try-on model
  • CatVTON: Integrate CatVTON image virtual try-on model
  • Video VTON: Integrate video virtual try-on models
  • 3D VTON: Integrate 3D virtual try-on models
  • Labels: intermediate, virtual-try-on, open-source, help wanted

Advanced Contributions​

For experienced contributors:

Video Generation​

  • VideoGenerator Interface: Design and implement unified VideoGenerator interface
  • SORA2 Integration: Integrate SORA2 video generation API
  • LUMA Ray Integration: Add LUMA Ray video generation adapter
  • Open-Source Video Models: Integrate video models from HuggingFace/GitHub
  • Labels: advanced, video-generation, help wanted

Agentic AI System​

  • Multi-Agent Framework: Design multi-agent system architecture
  • LLM Integration: Integrate LLMs for text analysis
  • VLM Integration: Add VLM support for image analysis
  • PDP Optimizer: Build automated PDP evaluation and optimization pipeline
  • Labels: advanced, agentic-ai, help wanted

Core Infrastructure​

  • Model Registry: Implement model registry system for easy model discovery
  • HuggingFace Integration: Create utilities for loading models from HuggingFace Hub
  • CLI Tool: Enhance CLI with ot generate, ot tryon, ot video, ot agent commands
  • Async Support: Add async/await support for API calls
  • Labels: advanced, infrastructure, help wanted

Contribution Areas by Priority​

High Priority (November 2025 - December 2025)​

  1. Image Generation Providers - Nano Banana, GPT-Image-1, FLUX Kontext Pro (closed-source), FLUX.1-dev (open-source)
  2. Core API Infrastructure - Base classes, interfaces, unified responses
  3. ImageGenerator Unified Interface - Standardize image generation adapters
  4. Testing - Unit tests, integration tests, >90% coverage
  5. Documentation - API docs, examples, tutorials

Medium Priority (January - February 2026)​

  1. Open-Source VTON Models - Leffa, DreamO, Voost, IDM-VTON, CatVTON, Video VTON, 3D VTON integrations
  2. Video Generation - VideoGenerator interface, SORA2, LUMA Ray integrations
  3. Agentic AI System - Multi-agent framework, LLM/VLM integration
  4. Model Registry - Model discovery and management system
  5. CLI Improvements - Enhanced command-line interface

Nice to Have (February - April 2026)​

  1. Additional Providers - Fashn AI, TryOnLabs, Qwen-Image
  2. Additional Datasets - DeepFashion, FashionGen
  3. Advanced Features - Async support, caching, batch processing
  4. Developer Tools - VS Code extension, Jupyter integration

How to Contribute​

  1. Find an Issue: Check our GitHub Issues for tasks labeled good first issue, help wanted, or intermediate
  2. Claim a Task: Comment on the issue to let us know you're working on it
  3. Fork & Branch: Fork the repo and create a feature branch
  4. Code: Write clean, tested code following our Contributing Guidelines
  5. Test: Ensure all tests pass and add new tests for your changes
  6. Document: Update documentation and add docstrings
  7. Submit PR: Open a pull request with a clear description

Contribution Guidelines​

  • Code Style: Follow PEP 8, use type hints, add docstrings
  • Testing: Write tests for new features, maintain >85% coverage
  • Documentation: Update README, add examples, document APIs
  • Commits: Write clear commit messages following conventional commits
  • PRs: Keep PRs focused, include tests and documentation

Issue Labels​

We use these labels to help you find tasks:

  • good first issue - Perfect for new contributors
  • help wanted - We need help with this
  • intermediate - Requires some experience
  • advanced - Complex tasks for experienced contributors
  • documentation - Documentation improvements
  • testing - Test-related tasks
  • api - API adapter work
  • image-generation - Image generation features
  • video-generation - Video generation features
  • virtual-try-on - Virtual try-on model integrations
  • open-source - Open-source model integrations
  • agentic-ai - Agentic AI system work
  • datasets - Dataset adapter work
  • infrastructure - Core infrastructure improvements

Get Help​

Current Focus Areas​

This Month (November 2025):

  • Dataset module testing and refinement
  • API adapter pattern standardization
  • Image Generation Model Integrations:
    • Closed-source: Nano Banana, GPT-Image-1, FLUX Kontext Pro
    • Open-source: FLUX.1-dev and other open-source models
  • Documentation improvements

Next Month (December 2025):

  • ImageGenerator unified interface
  • Model registry system
  • Core API infrastructure (base classes, interfaces)
  • Unified VirtualTryOn interface
  • Configuration management system

Coming Soon (January 2026):

  • Video Generation capabilities
  • Open-source VTON models (Leffa, DreamO, Voost, IDM-VTON, CatVTON, Video VTON, 3D VTON)
  • Agentic AI system foundation

Ready to contribute? Check out our Contributing Guide and start with a good first issue!