PulseAugur
EN
LIVE 06:42:02
日本語(JA) 動画生成AI「MiniMax H3」の開発チームが「画像生成・編集モデル」「動画アップスケールモデル」「低ステップ版MiniMax H3」を準備中であることを明言 https:// fed.brid.gy/r/https://gigazine .net/news/20260810-minimax-h3-team-ama

MiniMax H3 AI team plans new models, commits to open-source

The development team behind the open-source video generation AI, MiniMax H3, has announced plans to release several new models. These upcoming releases include an upscaling model, a lower-step version of MiniMax H3 for faster generation, and models for image generation and editing. The team also confirmed their commitment to keeping their models open-source, aiming to continue this practice until they achieve Artificial General Intelligence (AGI). Additionally, they are considering a transition to a more permissive license like Apache 2.0. AI

IMPACT This signals continued open-source development in advanced video and image generation, potentially accelerating research and adoption.

RANK_REASON The development team behind an open-source frontier model (MiniMax H3) announced future model releases and their commitment to open-source development. [lever_c_demoted from frontier_release: ic=1 ai=1.0]

Read on Mastodon — mastodon.social →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

MiniMax H3 AI team plans new models, commits to open-source

COVERAGE [1]

  1. Mastodon — mastodon.social TIER_1 日本語(JA) · [email protected] ·

    The development team behind the video generation AI "MiniMax H3" has clearly stated that they are preparing "image generation/editing models," "video upscaling models," and a "low-step version of MiniMax H3" https://fed.brid.gy/r/https://gigazine.net/news/20260810-minimax-h3-team-ama

    動画生成AI「MiniMax H3」の開発チームが「画像生成・編集モデル」「動画アップスケールモデル」「低ステップ版MiniMax H3」を準備中であることを明言 https:// fed.brid.gy/r/https://gigazine .net/news/20260810-minimax-h3-team-ama/