Pika releases open-weight video model Pika H3 Flex

Fri, Oct 30 01:08 UTC+8Updated Sat, Oct 31 20:55 UTC+8

Pika H3 Flex open weights are on Hugging Face: LoRA adapters (FL2VA, Ref2VA) on MiniMax-H3 that generate video with synchronized stereo audio in six sampling steps, with up to 12 reference inputs. Local inference is native 480p/768p, 5-15 s; 1080p, 2K and 4K output and separate audio stems come via Pika's hosted API. Licensed under the MiniMax H3 Community License.

Why it matters

Pika released H3 Flex, an open-weight video model built on MiniMax-H3 that generates 480p/768p video with audio locally; 4K and audio stems need the Pika API. Pika says it tops DesignArena's open-weight t2v board.

Sources2

Hugging FaceMediaFri, Oct 30 01:08 UTC+8Pika H3 FlexView original · Fri, Oct 30 01:08 UTC+8
@pika_labsXFri, Oct 30 01:37 UTC+8Meet Pika H3 Flex, debuting today as the #1 open weight model on DesignArena’s t2v board. It’s built for professional iteration, with exportable audio stems, resolutions up to 4k, and faster-than-realtime generation. It can produce a 15-second clip with soundtrack in as little as 7 seconds. So, yeah. It’s kind of a flex 💅View original · Fri, Oct 30 01:37 UTC+8

Written by dalao.ai: the title, summary and why it matters are our own words. The original reports are the record; their publishers keep authorship.