Pruna open-sources P-H3-Image: MiniMax-H3 video model turned into 8-step, 7x smaller local image model
cocktailpeanut · x · 2026-11-06
Pruna AI open-sourced P-H3-Image, an image generation and editing model built from MiniMax's SOTA video model H3. It generates 5x faster (8 steps instead of 50, 4 MP in 2.6s vs 12.4s) and is 7x smaller than H3, so it runs on much smaller GPUs.
- Renders 2-16 MP and edits with up to 8 input images; strongest at photorealism, documentary/editorial photography, film stills, and graphic styles with short quoted text
- PhospheneAI ported it to Mac: runs locally on Phosphene 4.19.19, free and open source, 8 MP without upscaling in 5 min on M4 Max (needs 48GB+ Mac and the Hailuo H3 pack, available via Pinokio)
- Pruna reworked every component of H3, including the prompt upsampler and text encoder
More from Infra
- Google to spend $17B on massive AI data center in Wyoming — Polymarket · 2026-11-06
- Hugging Face burns 800K GPU hours to release free 70K-word MoE training handbook — Thom_Wolf · 2026-11-06
- Coherent revenue jumps 50% to $2.37B, signs ~$20B five-year PhotonLink deal — BenBajarin · 2026-11-06
- Chip designer pitches Apollo on small AI data centers across 20 US states — pstAsiatech · 2026-11-06
- SpaceX data center pledges 40% power discounts and billions in community benefits — Scobleizer · 2026-11-06
- Mystery seller seeks to dump $500 million of data center loans — dinabass · 2026-11-06