MiniMax releases H3-Image: 4K image generation in under 10 seconds
iamfakhrealam · x · 2026-11-14
MiniMax launched H3-Image, an open-weight image generation model built on H3. It generates 4096×4096 4K images in under 10 seconds (claimed up to 5x faster than mainstream alternatives), supports up to 16 reference images, and improves realism and prompt adherence. The gains come from across the post-training pipeline: data, RL, sampling-step reduction, PE and deployment optimization.
Related event: MiniMax Open-Sources H3-Image for 4K Generation in 10 Seconds(2 posts)→
More from Multimodal
- Reddit user builds a 6-minute Caesar-at-Alesia animation in Google Flow with consistent characters — EpicFoolss · 2026-11-14
- Open-source Blender add-on runs NVIDIA Kimodo, ARDY and MotionBricks locally for AI motion generation — bdsqlsz · 2026-11-14
- NeurIPS 2025 best paper: diffusion models start memorizing at ~300 steps per image — Mastbubbles · 2026-11-14
- Sci-fi film KILL/SAVE team releases "fake ads from 2045" made with AI video — killsavecom · 2026-11-14
- Stella & Shango Part 2: Viking-themed AI video short continues — stellaXaiXvip · 2026-11-14
- Fable 5.2 trick: one recolored 3D character populates every spectator NPC — chongdashu · 2026-11-14