MegaParts: Scaling Part-Aware 3D Generation to 300 Parts via Efficient Modeling
Shanghai-AI-Laboratory · hf · 2026-08-18
MegaParts scales part-aware 3D object generation to 300 parts using token-efficient vector-quantized part tokens and structured autoregressive sequence modeling. It utilizes long-context training to enhance efficiency.
More from Multimodal
- Qwen3.8 Multimodal Model Released in GGUF Format for Local Inference — HauhauCS · 2026-08-18
- Anime Fantasy Magic Battle Scene Created Locally with Stable Diffusion at $0 Cost — foxdit · 2026-08-18
- MiniMax H3 generates multi-angle explosion scene with partial rewind effect — blackdatafilms · 2026-08-18
- FishAudio Beats ElevenLabs for Chinese Voice Cloning — yihui_indie · 2026-08-18
- PhiZero: Building a Predictable World Model with 'Physical Language' — 机器之心 · 2026-08-18
- blind_watermark: 14k-star open-source tool embeds invisible watermarks that survive resize, crop and blur — lxfater · 2026-08-18