Tencent-Hunyuan/HY-WorldPlay

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1586 stars · 143 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

HY-WorldPlay (HY-World 1.5) is Tencent Hunyuan's open-source framework for interactive 3D world modeling, generating explorable 3D scenes from images with real-time latency and geometric consistency. It includes pretrained WorldPlay models (5B and 8B variants), inference and training code, and RL post-training tooling.

Use cases

• generate explorable 3D worlds from a single image
• build interactive world models with real-time inference
• fine-tune or train a custom video-based world model
• run a world model on small-VRAM consumer GPUs
• apply RL post-training to a world model
• create 3D scenes for games or simulations from photos

When to choose

• you need interactive, real-time 3D world generation from images
• you want an open-source world model with training and fine-tuning code
• you have GPU resources and want geometrically consistent explorable scenes

When to avoid

• you need a lightweight CPU-only solution
• you need a fully permissive license (license is custom/unspecified)
• you only need static 3D asset reconstruction rather than interactive worlds

Facets

framework · maturity active

machine-learning deep-learning llm-inference simulation image-processing gpu-computing artificial-intelligence deep-learning simulation graphics python cross-platform world-model image-to-3d 3d-generation interactive-world-modeling video-generation geometric-consistency real-time-inference hunyuan tencent game-development gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="Tencent-Hunyuan/HY-WorldPlay")