Pika releases open-source H3 Flex video model, tops DesignArena t2v board at 4K
linoy_tsaban · x · 2026-10-30
Pika has dropped H3 Flex on Hugging Face, an open-weight text-to-video model the poster calls their favorite H3 fine-tune yet. Highlights: #1 open-weight model on DesignArena's t2v leaderboard, resolutions up to 4K, fast 6-step inference, and day-0 support in diffusers and ComfyUI. Weights and a demo Space are publicly available.
Related event: Pika releases H3 Flex, its first open-weight video model(19 posts)→
More from Multimodal
- Liquid AI launches LFM3 model family; 1.7B hits 185 tok/s on-device — JosephJacks_ · 2026-10-30
- Open-source ComfyUI film studio turns one prompt into an editable, refinable movie — Goat_bless · 2026-10-30
- Abiza turns a voicenote into a realistic AI influencer UGC video in seconds — adrianscottcom · 2026-10-30
- One-person AI music video: Suno + Hermes Agent + ComfyUI on a single RTX 3090 — Ok-Wolverine-5020 · 2026-10-30
- Same girl every scene vs. a new face each run: what a character LoRA actually changes — wuraang9 · 2026-10-30
- Projection mapper open-sources ComfyUI node pack that drafts full light shows locally on 12GB laptops — Puzzled_Parking2556 · 2026-10-30