PiD v1.5 Released: Fixes Color Bias and Enhances 4K Quality
bdsqlsz · x · 2026-07-14
PiD (Pixel Diffusion Decoder) v1.5 has been released, featuring two major improvements:
- Resolves previous color bias issues, delivering better 4K visual quality.
- Provides undistilled checkpoints and full training code for easy reproduction and secondary development.
The author notes support for models like FLUX, FLUX2, Qwen-Image, and Z-Image. Code, demos, and comparison pages are also provided to encourage users to directly utilize, reproduce, and build upon this work.
Related event: NVIDIA Releases PiD v1.5 Checkpoints(3 posts)→
More from Multimodal
- Getting Started with AI Video: Solving Consistency and Censorship — cynicalnewenglander · 2026-07-22
- Storyboard-first workflows are making AI dance videos and influencers more consistent — aftahi_ai · 2026-07-22
- Runpod MCP and Claude help spin up image and video generation workflows — 802high · 2026-07-22
- Midjourney prompt turns a bee into a glitching pixel explosion — michaelrabone · 2026-07-22
- A physics reward can improve video generation without creating a real physics engine — Dapper-Drawer4546 · 2026-07-22
- HeyGen adds a media-sourcing skill for coding agents with 75k images and 10k tracks — HeyGen · 2026-07-22